Collective Action and the Development of
Technical Standards in U.S. Industry
by
ALAN BRUCE DAVIDSON
S.B. Mathematics with Computer Science, Massachusetts Institute of Technology
(1989)
Submitted to the Department of
Electrical Engineering and Computer Science
In Partial Fulfillment of the Requirements for the Degree of
MASTER OF SCIENCE IN TECHNOLOGY AND POLICY
at the
Massachusetts Institute of Technology
June, 1993
C Massachusetts Institute of Technology, 1993. All Rights Reserved.
Signature of Author
,
-
/-HI
L
·
~-
~---
r
--
Department of Electrical Engineering and Computer Science
March 1, 1993
Certified by
Dr. Lee McKnight
-Thesi&Supervisor
Certified by
Dr. Richard Tabors
Thesis Supervisor
Accepted by
Professor Richard de Neufville
Chairman, Technology and Policy Program
ARCHivr-c
MASSACHUSETTS INSTITUTE
OF TFr('tNOLOGY
JUL 09 1993
LIBRARIES
Collective Action and the Development of
Technical Standards in U.S. Industry
by
ALAN BRUCE DAVIDSON
Submitted to the Department of Electrical Engineering and Computer Science
on March 1, 1993
in Partial Fulfillment of the Requirements for the Degree of
Master of Science in Technology and Policy
Abstract
This thesis seeks to answer the question, "Why are some industries better than
others at developing standards?" It is also an inquiry into why individuals get together to
do things that are in their common interest.
Standards and the processes through which they are developed serve a vital role in
the emerging global economy. Technical standards ensure the compatibility, quality, and
safety of products and systems. Billions of dollars are spent annually on the creation and
maintenance of standards, and their development is hotly debated by businesses and
governments.
The mechanisms for developing standards within an industry, the industry's
"standards infrastructure," can be thought'of as a collective good. The economic theory
of collective action provides a model for assessing when industries will invest in standards
development, by classifying industries based upon characteristics such as industry size,
market concentration, and the importance of standards.
The collective action model is tested here through case study observations of four
U.S. industries. The agricultural industry, a "latent" group with many small players, was
unable to develop standards through market mechanisms and now relies on the USDA to
set industry standards. The telecommunications industry, a "privileged" group with
fewer, larger players, exhibits a highly organized standards infrastructure. The medical
devices and machine tools industries, both "intermediate" groups, display differing results:
the medical device community has evolved highly effective organizations, while the
machine tools industry is suffering real competitive losses from its inability to develop
standards successfully.
The results of the case studies support the collective action model of standards
development in U.S. industry. This may have important implications for the ongoing
policy debate about the appropriate roles of public- and private-sector standards
initiatives. The collective action model predicts that market mechanisms will not always
provide adequate standards. The intervention of government, or the creation of other external incentives, may be required to promote an efficient level of investment in standards.
Thesis Supervisor: Dr. Lee McKnight
Title: Research Associate, Center for Technology, Policy, and Industrial Development
Thesis Supervisor: Dr. Richard Tabors
Title: Senior Research Engineer, Laboratory for Electromagnetic and Electronic Systems
Acknowledgments
This thesis, and the intellectual journey it symbolizes, would not have been possible
without the inspiration, ideas, and support of many. In particular, I would like to express
my thanks to:
The Hugh Hampton Young Foundation, and Mr. & Mrs. Bernard Rabinowitz, for their
financial support throughout my graduate career at MIT. Their help enabled me to attend
MIT; their generosity gave me the intellectual freedom to truly explore my interests in a
way that too few graduate students enjoy.
The U.S. Congress Office of Technology Assessment's International Standards Project, where
much of the background research for this thesis took place. I would especially like to
acknowledge a tremendous intellectual debt to Linda Garcia and Sherry Emery, Project
Director and Analyst respectively for the Standards Project. In the fiery crucible of
exploration, collaboration, and discussion involved with an intensive study, it is often
impossible to distinguish where thoughts begin and end. In that spirit, many of the ideas
expressed here are as much theirs as they are my own. Of course, any errors that remain are
the sole responsibility of the author.
Richard Tabors and Lee McKnight for their intellectual input and extraordinary patience
and understanding during an unusually arduous and extended writing process. They went
above and beyond the call of duty in accommodating me and coaxing me through a thesis
that took far longer than it should have.
The men and women of the Technology & Policy Program, for providing a wonderful
intellectual community that felt more like a home. Special thanks to Rene Smith and Gail
Hickey, den mothers and angels of mercy both, for their kind words and smiles.
Dean Guido Calabresi of Yale Law School, who said the right words to me at the right
time, to get the ball rolling on this project once again.
Frank and Izaline Davidson, whose home in Concord provided the ideal environment for
house-sitting and writing, and whose hospitality is matchless.
My family, to whom I owe everything.
And, finally, Andrea Shapiro, my old, dear friend and the editor of this thesis. In the final
weeks of this effort, her moral support and practical assistance were invaluable. I am not
sure I would have been able to finish if it were not for her.
Alan Davidson
New Haven, CT
February 26, 1993
Dedicatedto
Bruce andLeona Davidson
Robert andLuise Davidson
Adolph and Irma Kerpel
and to Andrea,
wherever I may find her
Contents
........... .................... 9
1. Intro d uctio n ........................................................................
1.1. The Problem ........................................
....................
10
1.2. Framing a Hypothesis .....................................
..................................
11
1.3 Outline...............................
14
..............................
2. A Framework for Analysis: Standards in U.S. Industry ....................................... 15
2.1. Standards: Definition and Function ......................................................... 15
2.2. The Importance of Standards ..............................................................
18
2.3 Standards-Setting Mechanisms.................................... 20
2.4 The Industry-Infrastructure View.........................
......
..................... 25
2.5 Prior Works and the Search for a Hypothesis ......................................... 26
3.
Developing a Model: Standards as Collective Goods ...................................... 29
3.1.
Standards Development as a Collective Good .................................... 29
3.1.1. Definition of Collective Goods................................................. 29
3.1.2. Standards as Collective Goods..................
.......
31
3.1.3. Efficiency and the Provision of Public and Collective Goods....... 33
3.2. The Theory of Collective Action ..................................
.................... 35
3.2.1. The Collective Action Dilemma ............................................... 36
3.2.2. Small Groups and Large Groups ..............................
........... 37
3.2.3. A Taxonomy of Groups: Privileged, Intermediate, and Latent
40
Groups.........................................
3.3. Applying Collective Action Theory as a Predictive Model for
..............................................
42
3.3.1. Predictions .......................................................
43
3.3.2. Potential Policy Implications............................................
44
Standards Setting...... ................. ...............
4. Industry Case Studies ............................................................................................
4.1. Introduction.........................................
46
............................................... 46
4.1.1. Selection of Cases .....
....... ........ ..........................
..
... 46
4.1.2. M ethodology...........................................................................
47
.......... 49
................................
....
4.1.3. Predictions
51
4.2. Agriculture ........................................
4.2.1. Agriculture in America ................................................. 51
4.2.2. Standards and Agricultural Products..................................... 53
4.2.3. Evolution of the USDA's Role ........................................... 55
4.2.4. Agricultural Product Standards Today.........................
57
4.2.5. Summary and Analysis: Agriculture and Collective Action......... 60
..
4.3. Medical Devices.................................
....................
63
4.3.1. The Medical Devices Industry.................................................. 64
4.3.2. Standards and Medical Devices...
.............................. 67
4.3.3. Summary and Analysis: Medical Devices and Collective
A ction .................................................
4.4. M achine Tools . ...
... ..............
................................. 71
........... ....................................... 77
4.4.4.1. Machine Tools in America ..............
......
4.4.2. Standards and Machine Tools ....................
..... 77
........................
.................... 79
4.4.3. Summary and Analysis: Machine Tools and Collective
Action ........................
. .........................................
4.5. Telecommunications .....................................................
83
86
4.5.1. Telecommunications in America ................................................ 86
4.5.2. Standards and Telecommunications ........................................... 88
4.5.3. Summary and Analysis: Telecommunications and Collective
Action ............................................
4.6. Summary of Results ........................................................................
5. Conclusion and Policy Analysis.....................
92
96
.......... ................................................ 97
5.1. The Policy Problem.....................................................................
97
5.2. The Policy Debate: Private or Public? .. ... .
. . . ................
..
.. ... . .
.. . . .
5.3. Implications of Collective Action Theory ..................................
5.4. Areas for Further Research ...................
....
105
.......................... .................... 108
...................
5.5. Conclusion......................
6. References............
99
.................................
110
.......... 112
"Men journey together with a view to particular advantage,
and by way of providing some particular thing needed for
the purposes of life..."
-- Aristotle, Ethics
1.
Introduction
Standards are all around us. From VHS videocassette recorders to IBM-
compatible personal computers, from USDA Grade A eggs to the 110-volt microwave
ovens we cook in, standards are a ubiquitous and vital part of the modem economy and
everyday life. As products become more complex, the need for standards increases. As
markets become more global, standards across international borders take on new
significance for competitiveness. Today, billions of dollars are spent annually on the
creation and maintenance of standards. Various mechanisms for developing standards, and
appropriate roles for the state in those mechanisms, are being hotly debated in corporate
boardrooms and the halls of government.
Some industries, such as the medical devices or chemical products industries, have
been particularly adept at making and using standards. Others, such as machine tools, have
seen their inability to create standards lead to real competitive losses. At the same time,
little theoretical literature exists which analyzes the successful development of standards at
the industry level.
This thesis seeks to answer the question "Why are some industries better than
others at developing standards?" On a more fundamental level, it is also an inquiry into
why people get together to do things that are in their common interest. The standards
infrastructure for an industry is a form of "collective good," the benefits of which are
available to the group as a whole. Existing economic theories about public goods and
collective action provide one method for assessing when industries will or will not be able
to produce collective goods, or in this case a standards infrastructure. This model will be
tested through a series of case studies on standards development in four U.S. industries.
1.1. The Problem
Technical product standards are agrt , ients that govern the functionality, quality,
or reliability of a product or process. They serve to guarantee that different products will
be compatible where necessary; that requirements of quality are met; and that basic levels
of health and safety are ensured. Standards have historically been an essential part of
modem economic development. Such notable innovations as standard railroad gauges, the
110 volt standard for home electrical outlets, and the standardization of auto parts are just
a few examples of how standards have played a substantial role in the growth of the
modem economy.
The development of effective standards within an industry requires the thoughtful
input of a broad variety of stakeholders, as well as a mechanism for collecting that input,
disseminating its results, and collectively acknowledging standards. This "standards
infrastructure" can be costly and difficult to establish and maintain, and requires a high
degree of organization.. Different industry sectors within the US have had differing
degrees of success at developing these mechanisms. This paper attempts to understand
those varying results by proposing a model for standards development based upon the
theories of collective action.
One caveat: it may appear that this analysis assumes that all standards are
desirable. This represents an over-simplification of a complex situation. The question of
whether or not a particular standard is desirable is a difficult one. While standards do
provide the benefits mentioned above, they can also: mistakenly adopt an inferior or
inefficient technology or process; lock in older technologies beyond their useful life, or
adopt immature new technologies too early; or create barriers to entry and concentrate
market power.'
This thesis does not attempt to develop normative measures of what a
"good" standard is. Rather, to the extent that participation in the standards development is
desirable, this thesis attempts to understand whether there will in fact be efficient levels of
investment and participation in the standards process.
1.2. Framing a Hypothesis
The standards process is on of suprising complexity and subtlety, yet little research
exists on the dynamics of the process within the United States. The available academic
literature tends to fall into three categories: 2
1.
Theoretical work based on game theory and microeconomics, often
accompanied by illustrative case studies. This research has tended to
concentrate on the adoption of standards by individual firms or consumers,
the participation of firms in standards development, and the theoretical
economic failures that arise in these areas.
2.
Conclusions based on case study research, often to validate prior hypotheses.
This work has focused almost exclusively on the development of particular
standards, rather than attempting to broadly consider standards mechanisms
within or across industries.
3.
Macroeconomic approaches using aggregated statistics to gain insight into
the development of standards.
Only the last of these specifically addresses the development of standards on an industrywide basis. The macroeconomic literature is limited, however, and provides no
'Stanley Besen and Leland Jolmson, Compatibility Standards, Competition, and Innovation in the
BroadcastingIndustry, Washington, DC: RAND Corporation, 1986. Walter Adams and James W.
Brock, "Integrated Monopoly and Market Power: System Selling, Compatibility Standards, and Market
Control," QuarterlyReview of Economics and Business, Vol. 22, No. 4 (Winter 1982) 34.
2As noted by Marvin Sirbu and Martin Weiss, "Technological Choice in Voluntary Standards
Committees: An Empirical Analysis," Economics ofInnovation and New Technology, vol. 1 -(Harwood
Academic Publishers, 1990) 111-133.
overarching theoretical framework for use. 3 This thesis differs from prior research, then,
both in its broad industry-wide scope and cross-industry perspective, and in its attempt to
provide a theoretical framework for considering the development of standards processes.
Collective action theory will be used as a model for the development of standards
in four industries: telecommunications, agriculture, machine tools, and medical devices.
Mancur Olson's seminal work in the field, The Logic of Collective Action, provides a
taxonomy of groups based on the size and number of the stakeholders involved. The
theory predicts that different types of groups will provide collective goods to different
degrees. Olson's predictions about the provision of a collective good by each type of
group will be compared with the provision of standards actually observed within each
industry.
The objective of this thesis is to show that the body of "collective action theory"
makes useful predictions about the development of standards processes, in a more
satisfactory way than current models based solely on market forces or government
regulation. Using case studies of industry standards processes, the predictions of
collective action theory are tested against observed results and compared with similar
analyses provided by other theories. The policy implications of these conclusions are then
considered.
Selection of cases: The four industries were originally selected to correspond to
the group-space set out by Olson. (See Table 1 below.) In addition, the industries chosen
all have , significant degree of international trade, so efficacy of standards in the
international arena could be considered. Standards play an important role in some aspect
of each industry, so standards development issues are relevant and important. The
3The
works of Link and Lecraw do support the hypothesis here, however, inthat they both observe that
highly concentrated market structures tend to produce more industry standards. Albert Link, "Market
Structure and Voluntary Product Standards," Applied Economics 15 (1983): 393-401; Donald J. Lecraw,
"Some Economic Effects of Standards," Applied Economics 16 (1984): 507-522.
industries chosen exhibit widely different standards mechanisms, degrees of coordination,
and perceived levels of success, to provide some basis for comparison.
Collection of data: Much of the case study data came from research done by the
author as part of the U.S. Congress Office of Technology Assessment's study Global
Standards:Building Blocks for The Future.4 Information about the cases comes from
both literature reviews and extensive interviews with standards participants in industry,
government, and academia.
Table 1. Olson's Taxonomy of Groups and the Industry Cases
Group
Few, large firms
Industry case
Telecommun-
Prediction
Standards provided
Initial observation
Highly organized
Standards not
provided
Disorganized,
reliance on
ications
Many, small firms Agriculture
government
Several, mediumsized firms
Machine tools
Medical devices
Unclear;
"Exploitation of the
large"
Machine tools:
Disorganized
Med. devices:
1Organized
Alternative models: Conventional wisdom within the standards community
conforms to two basic models. One promotes the ability of private-sector market
mechanisms to supply standards in an efficient manner. The other looks to differing
degrees of government involvement to maintain the standards process. These two
alternative models will be examined in the context of each case, with an eye toward
evaluating their usefulness relative to the collective action model. Other more sociological
explanations, such as the relationships and animosities between stakeholders in the
standards process, will also be considered; while it is likely that these explanations are too
4U.S. Congress, Office of Technology Assessment, GlobalStandards: Building Blocksfor the Future,
(Washington, DC: U.S. GPO, March 1992).
limited in scope to provide a generally useful model, they may resolve soi;e' of the
inconsistencies between theory and observation.
1.3. Outline
Chapter 2 - Standards in the U.S.: A Framework for Analysis - Chapter Two
provides a brief overview of the methods and organizations involved in setting standards
in the United States, as well as a discussion of the importance of industry standards
infrastructures.
Chapter 3 - Developing a Model: Standards as Collective Goods - Chapter Three
provides an analysis of standards as collective goods. It includes an overview of collective
action theory, as well as examples of its application to problems of standard setting in the
U. S.
Chapter 4 - Industry Case Studies - Chapter Four contains case studies of four
different U.S. industries which use standards, including an overview of standards
mechanisms within each industry and an evaluation of their development in the context of
collective action theory.
Chapter 5 - Conclusions and Policy Analysis - Chapter Five offers a summary of
the results of the industry case studies, in the context of the predictions made by collective
action theory. An overview of the standards debate is provided, with implications of the
results of this study as well as potential areas for future research.
2. A Framework for Analysis: Standards in U.S.
Industry
This chapter provides an overview of the standards setting environment in the
United States. The function and importance of standards in the nation's economy are
defined. Methods of setting standards are presented, as well as the major standards
organizations involved in the United States and internationally. Finally, the problems of
the standards setting process in the U.S. are considered, along with an overview of the
theoretical literature that explores these questions.
2.1. Standards: Definition and Function
The American HeritageDictionary defines standards as:
...An acknowledged measure of comparison for quantitative or qualitative
value; criterion;
...A degree or level of requirement, excellence, or attainment. 5
The U.S. Congress Office of Technology Assessment notes that "standards help determine
the efficiency and effectiveness of the economy, the cost, quality and availability of
products and services, and the state of the Nation's health, safety, and quality of life". 6
One useful way of looking at standards is to consider how they function. Marvin Sirbu has
noted that there are four basic types of standards: compatibility, information, variety
reduction, and quality standards.7
5American HeritageDictionary, Second College Edition (Boston,
MA: Houghton Mifflin Co., 1985),
1188.
60TA,
Global Standards, 7.
7Marvin
Sirbu, as cited in Lee McKnight, "The International Standardization of Telecommunications
Services and Equipment," in E.J. Mestmaecker, ed., The Law and Economics ofInternational
Teleco'mnunications(Baden-Baden: Nomos Verlag, 1987) 419.
Compatibility standards meet a recognized need for interoperability or
commonality in product form or function. For example, the system-wide compatibility of
railroad track gauges and boxcar axles is essential to the proper functioning of a longdistance rail nztwork. 8 Similarly, compatibility has been crucial in the development of the
phone system and modem data communications networks.
Variety reduction standards limit the versions of a product or process to simplify
understanding and maintenance, or achieve other economies. For example, standardized
automobile parts serve a dual role in streamlining manufacturing and allowing consumers
to have their vehicles serviced anywhere in the country. Even the colors of products are
standardized: Kenmore may produce a line of appliances in a particular color, so that if a
consumer needs to replace his washer, but not his dryer, he can be sure of finding one that
matches.
Information standards provide reliable ways of describing complex goods. For
example, a standard "Group 3" FAX machine lets the consumer know the capabilities of
the machine, without describing the encoding technique, handshaking protocols, or error
checking methods in their full technical detail.
Finally, quality standards serve as broad guarantees of a products reliability, safety,
or other characteristics of worth. Examples might include the consumer product safety
standards for children's toys, or the reliability standards for light bulbs or car brakes.
Another way of looking at the world of standards is through the role they play in
society. The Office of Technology Assessment, for example, has broken down the
standards universe into product standards, process standards, and control standards. 9
8Witness
the great railway standardization battles in nineteenth century Britain. Different railway gauge
standards were also used as a defensive mechanism by countries in Europe; new track sizes at the border
might have interrupted the flow of commerce, but it also prevented foreign troops from using rapid rail
transport. Charles Kindleberger, "Standards as Public, Collective and Private Goods," Kyklos, Vol. 361983-Fasc. 3, 386-387.
9 OTA,
Global Standards, 99-101.
Product standards serve as a kind of shorthand to both consumers and producers.
These standards "save time indescription, allow for co. ,entional discounts or premiums
for divergence from the standard, and permitting trading over distance without
inspection".' 0 They provide information about qualities of a product, without spelling out
the full details; they also serve as guarantees of those qualities once the standard has been.
For example, the public can expect that toothpaste approved by the American Dental
Association meets certain minimum efficacy requirements; doctors who purchase medical
devices sanctioned by the American Medical Association can rely on the safety and
reliability of the equipment. When bicyclists purchase helmets, they look for the Snell seal
guaranteeing crash performance; and many electrical appliances display the Underwriter's
Laboratory seal of approval, a certification of safety for use inthe home.
There is a great degree of public trust implicit in the use of these standards. Many
merchants and consumers are unaware of the processes involved instandards development
and testing, or even what the standards mean. Yet they still may only accept products that
meet those standards.
Process standards "define roles and relationships, establish the rules for
interpreting behavior, and specify the way inwhich a particular procedure is executed"."
One example is the use of standardized forms; these forms provide an accepted, organized
way of exchanging information, as opposed to the chaos of individual information
gathering.12 The procedure for communicating school or business closings during a snow
emergency is another form of process standard: a "phone pyramid" may be established, or
certain radio stations may carry the broadcasts of closings.
1'Kindleberger, 378.
I"OTA, Global Standards, 100.
12James
R. Beniger, The ControlRevolution: Technology andthe Economic Originsof the Information
Society (Cambridge, MA: Harvard University Press, 1986), p. 15-16.
Finally, control standards are "designed to address a societal problem or hazard".' 3
These standards are often regulatory in nature and serve to protect public health, safety,
environmental security, etc. For example, the Food and Drug Administration does
extensive testing before approving products for sale to the public. The Occupational
Health and Safety Administration promotes standards which will assure safety in the
workplace. The Environmental Protection Agency's strict automobile fuel efficiency
requirements are another example of control standards.
2.2. The Importance of Standards
The function and goals of standards give some indication of their importance
within the fabric of social organization today. For the individual or organization standards
fulfill an important need by specifying the quality, compatibility, or safety of a product or
system. There is also an increasing national stake in effective standard-setting.' 4
The national stake in standards reflects their growing importance to the economic
well-being and social goals of the country. The increasing use of private and government
standards as a means of regulating the public health and safety requires a new attention
from government. For example, the EPA and OSHA have developed nearly 8500
standards dealing with safety and environmental issues. 15 Particularly in areas where the
individual citizen has little opportunity to be involved in the process (for reasons of
expense or practicality), and especially where the government has delegated its regulatory
authority to the standards process, there is an obligation for the state to represent the
citizenry.
13 OTA, Global Standards, 100.
141bid., 7.
15Toth,
Robert, Toth Associates, (ed.), "Standards Activities of Organizations in the United States",
(Washington, DC: NIST Special Publication 806, February 1991), p. 3.
Moreover, there is renewed recognition of the importance of standards to the
national economy. In addition to serving important functions for businesses and
individuals, standards can increase the economic efficiency of the markets. 16 Moreover,
many countries are now realizing that standards can be used as an important strategic,
competitive tool. For example, the Germans and Japanese have been particularly proactive
in having there standards adopted abroad, to the advantage of their manufacturers a big
boost. 17
A number of trends are expected to exacerbate the need for standards in the
coming decade. These include:18 the trend towards a global economy where America no
longer plays a dominant role; greater state involvement in promoting national economies,
with standards as an important tool for industrial policy; the growth of multinational
corporations and other transnational groups; the changing nature of manufacturing
methodologies, which require a new level of sophistication in standards and integration;19
and the emergence of an information-based society. 20
The importance of standards may seem to imply that the creation of standards is
good. However, the notion of "success" in standards-setting can be an elusive one. The
question of when a particular standard is a "good" one is the subject of great debate. The
16Lecraw, 509-510. Standards can increase the efficiency of the market by reducing information
assymetries, internalizing certain externalities, and promoting compatibility and trade. However, they
can also be used to increase the market power of an existing, influential market player and and erect
barriere to entry in a market, reducing efficiency.
170TA, GlobalStandards, 78. See also, Chapter 4.4, Machine Tools Case Study.
8sI am indebted to Sherry Emery, Urban Studies and Planning Department, University of North Carolina
at Chapel Hill, for her insights in this area.
19Consider the "flexible specialization" of manufacturing discussed in Michael J.Piore and Charles F.
Sabel, The Second Industrial Divide (New York: Basic Books, 1984) 286-295.
are expected to play an especially important role in the global information economy's
expanding use of information, computer, and communications system. See William J. Drake and Lee
20 Standards
McKnight, "Telecommunications Standards in the Global Information Economy: The Impact of
Deregulation and Commercialization," Promethee, March 1988.
adoption of a standard does not always lead to efficiency or social good. For example,
standards can adopt an inefficient technology or process.2" They can lock inolder
technologies beyond their useful life, or adopt immature technologies too soon.22 They
can also be a powerful weapon for companies with market power who wish to erect
barriers to entry or perpetuate the concentration of the market.23
Perspective complicates the notion of "good" standards. What may appear to be a
good standard from a technical standpoint may turn out to be a poor business decision or
an inefficient economic move. What may appear to be a good technical and economic
solution may turn out to be bad from the perspective of broader public interests such as
health, safety, and the environment. Finally, what may appear to be a good standard from
the perspective of one country may prove detrimental to another.
For all of these reasons, the development of a normative framework for evaluating
a particular standard is difficult. This thesis, however, will concern itself mostly with the
mechanisms for creating standards and the level of participation by stakeholders in those
mechanisms. Some assumption is made that, to the extent that standards are useful,
participation in the standards process is desirable. It should be noted that widespread
participation inthe process by many stake holders could mitigate the adoption of
inefficient or unsuccessful standards by more fully incorporating the values of the
populous in the standards process.
21Consider,
for example, the de facto market adoption of the IBM-PC standard inthe early 1980s. While
the adoption of some standard was probably vital to the development of the personal computer market,
the arguably "low-tech" IBM standard prevented the market adoption of more advanced capabilities
available in other non-standard machines such as those made by Apple, Commodore, or Next.
22The debate over high-resolution systems standards in the U.S. may provide an example of both of these.
The inertia of the U.S. television broadcasting standard has certainly inhibited the market's use of
available, higher resolution media technologies. At the same time, in the late 1980s the standards
community narrowly avoided adoption of an analog, non-extensible standard in favor of the arguably
superior extensible digital standards under consideration today.
23See
Yale M. Braunstein and Lawrence J. White, "Setting Technical Compatibility Standards: An
Economic Analysis," The AntitrustBulletin, Summer 1985.
2.3. Standards-Setting Mechanisms
Standards are developed through one of three methods: de facto, de jure, or
voluntary consensus standards processes. De facto standards are set in the market,
through the process of supply and demand. The acceptance of VHS videotapes over the
Beta tape format, or the competition between IBM and Apple computers in the market,
are examples of de facto standards at work.
A second way of setting standards is through the voluntary consensus process.
Voluntary consensus standards are generally developed through the active participation of
interested parties. Participation in the development and use of the standard is voluntary;
the processes used are expected to be somewhat open to participation by all interested
parties, and some form of consensus decision making is used in group decisions. Often a
major standard setting organization, such as the American National Standards Institute
(ANSI) will certify the process. 24
De jure standards are usually set by government, with the force of law behind
them. Agencies such as OSHA, the FDA, and the FCC all use mandated de jure standards
as a means of regulation. Often these result when the market fails to successfully establish
standards on its own, or when decisions in the marketplace fail to incorporate safety,
environmental, or other social externalities. 25
24Note
that the degree of openness and consensus used isoften a source of controversy. For example,
ANSI and another major standards certifier, the ASTM, differ in this respect: ANSI requires a
consensus (i.e. unanimous) vote, but does not require open meetings; ASTM requires meetings open to
all, but allows for majority-rule voting. OTA, Global Standards, 12-13.
25 When efficient standards do not emerge in a timely fashion, the market is said to have failed: "Ina
well functioning market, economic relations are governed by self interest, so it is self interest that drives
outcomes.... The market is considered to work well when it maximizes the goals of effieciency and
economic growth." OTA, GlobalStandards, p.101.
In the United States, standards are set largely through the private-sector voluntary
consensus method. This practice is attributed in large part to the American preference for
private industry. 26 The voluntary consensus process usually takes place through organized
committees staffed by individuals from interested parties, including industry experts,
academicians, and government representatives. The committees are usually sponsored by
large standards-setting organizations, professional societies, government, or ad-hoc
consortia.
Today, several large organizations are responsible for creating or approving the
bulk of standards in America. The largest of these organizations are the American
National Standards Institute (ANSI), followed by the American Society for Testing
Materials (ASTM), the American Society of Mechanical Engineers (ASME), and the
Institute of Electrical and Electronics Engineers (IEEE). They often play one of several
rules, including sponsorship of standards efforts, coordination of activities, or certification
of standards or standards processes sponsored by others.
27
These organizations also
receive the bulk of their revenue from the sale of standards.
The American National Standards Institute (ANSI) serves the primary (but not
exclusive) role of certifying these processes on the basis of certain accepted rules of
fairness and openness. Standards created by committees, organizations, or other groups
under ANSI guidelines can be then be designated "American National Standards."2 8
260TA,
GlobalStandards,46. As Alexis de Tocqueville points out in Democracy in America:
"Americans of all ages, all conditions, and all dispositions constantly form associations..."
"...Wherever at the head of some new undertaking you see the government of France, or a man of
rank in England, in the United States you will be sure to find an association".
1963 ed., pp. 106 and 110.
-
27
0TA, Global Standards, p. 13.
28"ANSI
sprinkles the holy water on the standards process." Anthony Bratkovich, NMTBA, interview
with author, McLean, VA, 10 July 1991.
The creation of standards through the committee process can be a long and
arduous task. The process usually consist of periodic working group meetings, where
experts from as wide a range of interests as possible try to work out the best possible
standards agreeable to all. Standards are usually approved by the body, then approved by
ANSI should they meet the right test. The whole process is also costly, requiring real
investment in terms of manpower, as well as expenses for mailings, copying, and frequent
meetings.
The Government also participates in the standards settings arena, generally as
more of an observer than leader or even participant. "Instead of setting standards for the
U.S. private sector, the government focuses its efforts on the fairness and effectiveness of
standards setting processes. '"29 The Government's role isbest articulated by OMB
Circular A-I 19, passed inthe early days of the Reagan administration, which requires the
federal government to rely on voluntarily, commercially-set standards whenever possible.30
The 1979 Trade Act also formally recognizes the private sector's role in standards
setting.31
Today, the government often finds itself a de facto standards setter; as a
monopsonistic buyer in many areas, the government's self-imposed standards process is
not open to outside players but still exercises inordinate power over the market. For
example, Department of Defense Military Specifications (MilSpecs) on everything from
the thickness of a military blanket to the tolerances ofjet engines have often become de
facto standards.
29
0TA, Global Standards, p. 15.
30US
Office of Management and Budget, "Federal Participation in the Development and Use of Voluntary
Standards," OMB CircularA-119(Washington, DC: GPO, 1982).
31Suzanne Troje, Office of the US Trade Representative,
interview with author, Washington, DC, 12 June
1991.
Internationally, the bulk of standards are set through large, voluntary consensus
standards groups. These include the International Organization for Standardization (ISO)
and the International Electrotechnical Commission (IEC),which oversee standards
committees and working groups on a wide variety of issues. Each country is represented
by a member group. In most countries that member is the national standards organization;
in the U.S. ANSI is the designated U.S. representative to ISO and IEC. The United
States is also involved in some standards organizations based on treaty, such as the
International Telecommunications. Union (ITU) and the Alimentarius Codex (for
agricultural products.)
In addition to the major consensus standards bodies, which administer
internationally recognized standards, each country has its own standards-setting groups
that must be dealt with. In contrast to their American counterparts, these groups tend to
be quasi-governmental in nature and often directly funded by the state. In Germany, for
example, the Deutsche Institute fur Normung e. V. (DIN) is formally recognized by the
German government as the pre-eminent standards body in that country; "the importance
of this orientation is that it allows the state to facilitate action by other actors which
promotes the achievement of long term national goals." 32
The international picture is complicated further by the rise of regional standards
organizations, such as the Comite Europeen de Normalisation (CEN), and the Comite
Europeen de Normalisation Electrotechnique. Sponsored by the EEC, they reflect the
European Community's recognition of the importance of standards harmonization as a tool
for union -- and as a powerful economic force. The growth of this complex web of
32Paterson,
William and Coli, Whitston, "Government-Industry Relations in the Chemical Industry: An
Anglo-German Comparison," Stephen Wilks and Maurice Wright (eds.), ComparativeGovernmentIndustry Relations (Oxford: Clarendon press, 1987) 38.
organizations and players makes effective representation in the international community a
difficult and costly affair. 33
2.4. The Industry-Infrastructure View
As noted inChapter 1, most of the theoretical literature on standards has
concentrated on how or why particular standards are adopted, and the impacts of those
decisions. 34 Another approach, particularly useful from the perspective of the policymaker, is the to look at the health of the standards infrastructure that has developed within
different industries.
The organized development of standards for an industry over time requires a high
degree of coordination and cooperation. Committees and working groups must be created
and publicized, expert personnel identified and brought together. Processes must be
agreed upon and approved. Within a large field with many standards, areas of
responsibility must be divided up. Sometimes just figuring out what standards need to be
made is difficult. Once standards are created, methods must be developed for their
dissemination and certification of compliance.35
33The
cost of travel alone to keep up with the latest developments and participate in meetings all over the
world several times a year can be exorbitant. For example, one ISO Motion Picture Expert Group
(MPEG) participant estimates that his company spends approximately $7,000 per year to send him to
meetings annually, plus three weeks of his time. Joseph Stampleman, Apple Computer, telephone
interview with author, February 28, 1993.
34
See, for example: Besen and Johnson; Marvin A. Sirbu and Laurence E. Zwimpfer, "Standards Setting
for Computer Communication: The Case ofX.25," Sloan School of Management, MIT, 1984;Martin
B.H. Weiss and Marvin Sirbu, "Technological Choice in Voluntary Standards Committees: An
Empirical Analysis," Econ. Innov. New Techn., Vol. 1, 1990, 111-133. (technical choice in the modem
market); or Joseph Farrell and Garth Saloner, "Installed Base and Compatibility: Innovation, Product
Preannouncements, and Predation," The American Economic Review, Vol. 76, No. 5, December 1986.
(network externalities, effects of excess inertia in standards or excess momentum).
35The issues associated with certification are as
complex and important as those of standards setting, and
have received are great deal of attention particularly within the context of international trade.
Unfortunately, they are beyond the scope of this inquiry.
The development of such a "standards infrastructure" within an industry is
important for effective standards. The timely creation and dissemination of standards is
facilitated by existing processes. People know where to turn when they need a new
standard developed; people also get to know where to turn when they need to find out
what a standard is. The marks associated with standards created by known organizations,
such as "American National Standards" or "ISO Standards," is in itself often crucial to
their adoption.
It makes sense to look at infrastructures along industry lines because the bulk of
stakeholders served by a particular standard infrastructure are usually concentrated within
an industry. While standards organization such as ANSI or ASTM do cross industry lines,
the certified committees and subcommittees where the actual work takes place are usually
administered and sponsored by industry trade groups and professional societies. That is
not to say that there are brig!- lines to be drawn between industries; telecommunications
standards have received their share of attention from other communities, for example. But
over time the bulk of standards activities, and the majority of beneficiaries, are organized
within industry boundaries.
The industry standards infrastructure, then, can serve as a useful basis for analysis
of the standards development process. A look at different U.S. industries where standards
are important reveals a striking difference in standards infrastructures: some, such as the
chemical products, telecommunications, or medical devices industries, have highly
organized standards processes. Others, such as machine tools or the agriculture industry
before USDA intervention, have had great difficulty in organizing effectively to create
standards. Understanding this dynamic is the problem to be addressed in this thesis.
2.5. Prior Works and the Search for a Hypothesis
Very little of the existing literature addresses the issue of standards processes on
an industry-wide level, although a few basic empirical studies provide a starting point for
this analysis. The prior work falls into three general categories: theoretical
microeconomics analyses, case studies, and macroeconomic surveys.36
The theoretical work tends to be based on game theory and microeconomics, often
accompanied by case studies. Most of the works explore the economic impact of choices
in standards, or the decisions made by firms in adopting standards. Prominent examples
include: Saloner & Farell, on the problems of excess inertia, momentum, and other
externalities that influence choices in standards setting; Katz & Shapiro, on the impact of
network externalities; and Braunstein & White, on the broad economic and social impacts
of standards.3 7 Most of this work does not look at the development of industry-wide
mechanisms as a phenomenon.38
The case-study work has tended to concentrate on the development of particular
standards, rather than standards-setting in an industry over time. Prominent examples
include the work of: Besen and Johnson, on the creation of various broadcasting
standards; Sirbu & Zwimpfer and Sirbu & Stewart, on factors in the selection of the X.25
communications standard; and Weiss & Sirbu, on factors in the choices made by voluntary
standards committees. While this work often considers factors such as industry size and
structure, it has tended to focus on technological choices in particular standards rather
than the development of standards mechanisms.
The macroeconomic surveys conducted by Link and by Lecraw are most relevant
to this study. These works have attempted to use highly aggregated industry data to make
inferences about the provisioning of standards. Link has suggested a correlation between
36Weiss and
37Saloner
Sirbu, 112.
and Farell; Braunstein and White; Michael L.Katz and Carl Shapiro, "Network Externalities,
Competition, and Compatibility," American Economic Review 75 (3)(1985): 424-440.
38The work of Braunstein and White does look at industry structure, but concentrates on the role of the
dominant firm in standards choices rather than industry mechanisms. Sanford Berg also considers the
standards as public goods, but again addressed only its impact on technical choices of standards; Berg
also focused on demand-side incentives, rather than producer incentives to standardize.
the market concentration and the presence of standards in an industry. Lecraw's work
linked the market power of buyers and the presence of standards for a given product. The
two of them taken together form the basis for analysis of industry standards based on
industry structure and concentration. 39
Lecraw touches on an important point about industry standards mechanisms: they
exhibit the characteristics of what economists call "public" or "collective goods."4 These
are goods whose benefits are not "appropriable" 'yany single user, but confer to a group
as a whole. The special problems involved with providing collective goods are the subject
of a good deal of economic literature. This body of "collective action" theory in turn can
be used as the basis of a model for understanding the provisioning of industry-wide
standards infrastructures.
39Their
work did differ from this thesis in some important ways. Lecraw's work analysed a number of
factors, looked at buyer's market power, and considered standards primarily at the product level. Link
focused on the market concentration and the presence of a single standard as an indication of standards
mechanism. Neither provided an over-arching theoretical basis for their findings, although both noted
the public goods characteristics of standards in these contexts.
40Lecraw, 509.
3.
Developin2 a Model: Standards as Collective
Goods
Industry-wide product standards can be thought of as "goods" which are costly to
produce but whose benefits are appropriable by a wide group of stakeholders. As such,
they have the characteristics of what economists call "public goods" or "collective goods."
A significant body of literature exists exploring public and collective goods and their
implications for public finance. Specifically, collective action theory provides a framework
for understanding when a market system will or will not supply efficient quantities of a
collective good. This framework can be used to make predictions about the provision of
standards, based upon such factors as industry structure and the importance of standards
within the industry. It may also indicate prescriptive measures that can be taken in cases
where only sub-optimal levels of standards are provided. It is this predictive model that
will be tested in Chapter 4. Other factors that may or may not influence the success of
standards mechanisms will also be identified.
3.1.
Standards Development as a Collective Good
3.1.1. Definition of Collective Goods
Public goods include those goods the benefits of which may be enjoyed by one
consumer without reducing the benefits enjoyed by other consumers of that same good. 41
Examples of "pure" public goods include national defense or major public works projects
such as uncongested bridges or roadways. "Impure" public goods, such as public
4 1Edwin
Mansfield, Microeconomics: Theory andApplications, 5th ed. (New York: W.W. Norton &
Company, 1985), 491.
education, exhibit many of the characteristics of "pure" public goods but are not wholly
public or appropriable. It has long been recognized that market mechanisms alone will not
induce consumers to reveal their preference for these goods, and will fail to provide these
goods at optimal levels.42 Non-market, or special mechanisms, such as government or
charitable intervention, are necessary to provide such goods in optimal quantities. 43
Collective goods are a more general case of public goods, where benefits are
conferred to a limited group rather than the population as a whole. A collective good can
be defined more precisely as an economic good "such that, if any person Xi in a group X1,
..., Xi , ..., Xn consumes it, it cannot feasibly be withheld from the others in that group."44
Collective goods share two overlapping, defining properties of public goods: 45
1. Nonexcludability -- Traditional microeconomics theory assumes that those who
pay for a good can consume it, and those who do not pay for a good cannot. For
collective goods, individual consumers within the group cannot be kept from
enjoying the benefit of the good once it has been provided, even if they have not
paid.46 If the good is available to one member of the group, then its benefits are
42 Mark
Blaug, Economic Theory in Retrospect (Cambridge: Cambridge University Press, 1985), 218,
596.
For example, see Paul A. Samuelson, "The Pure Theory of Public Expenditures," The Review of
Economics and Statistics36 (1954): 387-389. Public goods began to receive attention from economists
as early as the 1890s in the writings of Ugo Mazzola, and John Stuart Mills alluded to their existence
even earlier in the century.
43Mansfield,
493-501.
44 Mancur Olson, The Logic of CollectiveAction: Public Goods andthe Theory of Groups (Cambridge,
Mass.: Harvard University Press, 1971), 14.
45This definition is a distillation of several more lengthy discussions of public and collective goods. See
Mansfield, 490-491; Blaug, 596; Olson, 14-15; Agnar Sandmo, "Public Goods," in The New
Palgrave:A DictionaryofEconomics, ed. J. Eatwell, M. Milgate, and P. Newman, (London:
Macmillan, 1987), 1061-1065.
4 6Olson's
discussion of collective goods considers this to be the essential feature of a public or collective
good. He goes on to distinguish between exclusive collective goods - those which do exhibit rivalry of
consumption - and inclusive public goods -those whose use does not detract from their value to other
consumers. For the purposes of this discussion, "collective good" will be synonymous with Olson's term
"inclusive collective good." Olson, 37.
appreciated by all members of the group. National defense is one example of such
a good: once it has been provided, everyone in the country benefits from it.
2. Nonrivalry (iointness of supply) -- Traditional microeconomics also assumes that
the more a good is consumed, the less it is available for others to consume. For
example, if X drinks a beer, Y cannot then drink that same beer. For collective
goods, consumption by one consumer does not prevent another consumer from
also enjoying the good; one individual's use of a bridge does not mean that another
cannot use it (so long as the bridge is not too crowded.)
Public goods analysis tends to emphasize the latter of these characteristics, collective
goods analysis the former. This may be because the public goods literature is more
concerned with public finance and social welfare implications, where jointness of supply is
an indicator of waste and inefficiency. Collective goods literature places more emphasis on
group behavior and the free-rider problems associated with nonexcludability. For the
purposes of this analysis, collective goods will be considered to exhibit both of these
characteristics, with non-excludability as the crucial feature. In fact, the extent to which
the collective goods exhibit both of these characteristics and to which the "collectiveness"
of the goods can be limited will have important policy implications.
3.1.2. Standards as Collective Goods
Standards development processes are economic goods. The development of
standards has a significant real cost, in terms of both money and personnel. This cost
exists not only in the development of specific standards; rather, each individual
development effort takes place within the larger context of industry-wide standards
development processes. Within the definition of nonexcludability and nonrivalry, the
mechanisms and processes that an industry employs to develop technical product
standards -- the organizations, resources, personnel, and other infrastructure devoted to
development or dissemination of standards -- are collective goods for the group of
standard users.
Nonexcludability -- An effective industry standards development process, which
produces good standards in a timely manner and represents U.S. industry positions in the
international arena, will tend to benefit every standards user within the industry (in this
case, the collective group). This is tnre regardless of whether the standards user has
supported the process or contributed to its cost. The benefits of the existence and
maintenance of an effective standards infrastructure accrue to the industry's standards
users as a whole.
In some instances, proprietary or patented standards may keep some consumers
from enjoying the benefits of the standards process. In such cases, to the extent that there
are still "standards" being developed and not just competing products, the process
becomes a private good for some and a collective good for others. It remains a collective
good for some since inmost cases the benefits of standards will accrue to a broad
spectrum of users who benefit purely from the existence of a standard, regardless of what
the standard is.47
Of course, in some cases efforts have been made to exclude those who did not aid
in the development of a standard from using that standard. In cases where measures
independent of the development of the standards themselves are used to exclude noncontributors, there is infact a form of outside "coercion" taking place regarding provision
of the good. This "coercion" will be discussed later in this Chapter and in the policy
options in Chapter 5. In general, we will consider standards to be in the public domain
47For example,
consider a networking standard development process that chooses a patented standard.
While participation in the process may become a private good for those manufacturers competing for
the standard, the process itself remains a collective good to all of those end users who benefit purely
from existence of an agreed upon standard. These consumers cannot be kept from enjoying the benefits
of existence ofan effective process for developing standards, if they are willing to pay the licensing fee,
e.g., MS-DOS compatible computer.
once adopted, standards processes to provide benefits to both contributors and noncontributors.
Non-rivalry -- Once standards development mechanisms and standards have been
provided, their use by one industry group member does not detract from the benefits
accrued from their use by another consumer. 48 On the contrary, in many cases
consumption of the good tends to increase the value of the standards mechanisms.
Widespread recognition of a standards-setting body and the standards it creates is essential
to its value and success.
It can be seen then that standards development mechanisms do indeed fit the
definition of collective goods. They are costly to provide; their use by one consumer does
not detract from their value to other consumers, and typically enhances it; and it is not
possible to directly prohibit any member of the group from enjoying the benefits of the
process. Some standards development processes have incorporated measures to mitigate
their "collectiveness" (and it will be seen that this can be an effective tool in ensuring the
efficient provision of standards). However, the standards development process is
inherently a collective good.
3.1.3. Efficiency and the Provision of Public and Collective Goods
As noted above, economists have acknowledged for some time that special
mechanisms, external to the market, are required if public goods are to be provided in
optimal quantities. Forcing consumers to pay for a public or collective good in order to
48
1t should be noted that in certain special cases this may not hold absolutely. In some cases it might be in
an organization's interests to have a proprietary technology adopted as a de facto standard, but not
adopted as a standard by their rivals. In such a case, the technology is not really a standard in the
narrow sense defined in Chapter 2, but rather is just one technology competing for adoption. In most
cases, any group member that benefits from the development of standards will benefit by the
acknowledgment and use of those standards mechanisms by other members of the group.
use it is not efficient because -- in the case where there is jointness of supply -- excluding
those who do not pay does not increase the satisfaction of others.4 9 Charging a fee will be
effective only for those who use the good. Those who find the cost imposed higher than
their demand for such will not pay. They will either accrue no benefit, or do so at the
expense of the others. Inherently, a "free-rider" problem exists for goods with
nonexcludability. A consumer cannot be barred from using a good once it is provided, and
cannot be forced to pay for the good in order to use it. For these reasons, even
conservative neo-classical economists acknowledge the need for extra-market mechanisms
to promote public goods.50
The collective action of groups with common interests has long been taken for
granted within classical economic theory. It is assumed that if it is in the profit-maximizing
interest of a group of consumers to band together to achieve some common goal, they will
do so. Moreover, the interest group theory of political action has gained a prominent role
in political science. These theories indicate that groups will act to provide a collective
good they commonly seek.
To the extent that standards processes represent the collective interest of a small
group of stakeholders acting to maximize their utility, the market should act to provide
this group with the activities it needs. However, because standards development
possesses the characteristics of a public good, it might also be expected that reliance on
market mechanisms would just as likely fail to provide effective standards development
processes, and that some sort of third party intervention would be required. This
49This
notion of efficiency isbased upon Pareto-optimality: "... society should make any change that
harms no one and improves the lot of some people. If all such changes are carried out - and thus no
opportunity to make such changes remains - the situation istermed Pareto-optimal or Pareto-efficient."
Mansfield, 460. Contrast this with Rawls' argument that "all social values ... are to be distributed
equally unless an unequal distribution ... isto everyone's advantage." J.Rawls, A Theory ofJustice,
(Cambridge, Ma: Harvard University Press, 1971).
SoSee, for example, Milton Friedman, Capitalism and Freedom (Chicago: University of Chicago Press,
1982), 23.
conundrum and seeming contradiction is addressed in a more detailed and thorough
discussion of the nwture of collective goods: Mancur Olson's The Logic of Collective
Action.
3.2. The Theory of Collective Action
Market economic theory generally assumes that rational individuals will act in their
own self-interest to maximize their own personal utility. An oft-assumed corollary to this
is the idea that individuals with common interests will act in concert to achieve those
common goals. In his 1965 work The Logic of Collective Action, Mancur Olson
attempted to provide an analysis of the problems of groups acting collectively to reach a
common goal or provide a commonly desired good. As he notes:
It is often taken for granted, at least where economic objectives are involved, that
groups of individuals with common interests usually attempt to further those
common interests. Groups of individuals with common interests are expected to
act on behalf of their common interests much as single individuals are often
expected to act on behalf of their personal interests. This opinion about group
behavior is often found not only in popular discussions but also in scholarly
writings... 51
Groups can most often be defined by the common goals of their members. The
organizations and associations prevalent in today's society form to promote their common
goals. These goals are in effect collective goods for the members of a group. In the
context of standards, the companies, individuals, and organizations that need and benefit
from the development of acknowledged standards form a group within this definition.
The mechanisms for developing these standards, including the resources devoted to
standards work, the associations formed to organize those resources and disseminate their
product, and the infrastructure required to support their efforts and gain acknowledgments
51
1Oson, 1.
as the industry standards, are a collective good sought by the group and benefiting its
members.
3.2.1. The Collective Action Dilemma
The key result of collective action theory is that "unless the number of individuals
in a group is quite small, or unless there is coercion or some other special device to make
individuals act in their common interest, rational, self-interestedindividuals will not act to
achieve their common or group interests."52 This argument rests largely on the realization
that the purpose of groups is to provide collective goods, and that collective goods are
often not provided in efficient quantities based on market mechanisms alone.
Consider the case of a group organized for a purpose; the achievement of this
purpose constitutes a collective good for the group. "Buying" the collective good -achieving the common end -- will have an economic cost, whether in money, time, or
manpower. Presuming that the individuals in the group each gain from the collective good,
and there is no rivalry or exclusion, each utility-maximizing individual actor should be
willing to contribute so long as the individual cost is less than the total individual benefit.
The optimal amount of the collective good would then be the level such that the marginal
aggregate benefit (achieved by all of the individuals in the group) was equal to the
marginal aggregate cost of the good.
Consider, however, the viewpoint of a utility-maximizing individual within a large
group, trying to decide how much to contribute toward the collective good. If the group is
large enough relative to the individual, and that individual's decision on whether to invest
inthe good or not will have little impact on the total amount of the good provided, there
will be no incentive for the individual to contribute anything toward the collective good.
52Ibid., 2.
The individual knows that whether or not he pays for any of the good, it will still be
provided at the same level. The individual will seek to attain the good at a level such that
the marginal cost of the good is smaller than the marginal benefit to the individual, or at no
cost whatsoever.
Thus a utility-maximizing individual within the group will rarely contribute
anything to the purchase of the collective good, even though it is worthwhile for him to do
so if everyone does.53 Only in the instance where providing some quantity of the good is
so valuable to the individual that he is willing to buy it himself, will an individual buy any
of the good. This is only likely for small quantities of the good, where the good is very
valuable to the individual, or where the individual is large relative to the whole group.
Otherwise, a rational individual will choose to buy only a small amount of a collective
good, and a group will not succeed in achieving its common purpose. This is the
collective action dilemma.
3.2.2. Small GrouDs and Laree Groups
It should be noted that in a small group there are more likely to be individual
actors who receive a large fraction of the benefits of the collective good. For these
individuals it will be worthwhile to buy some of the collective good themselves, regardless
of contributions from others. In this way, some amount of the collective good will be
53This has overtones of the classic Prisoner's Dilemma in economic game theory. Unfortunately, most of
the game theoretic analyses of standards are limited to analyses of the strategic decision by firms over
whether to adopt particular standards. See, for example, Farrell and Saloner, (December 1986). Amore
detailed game theoretic model of the actions of individuals on an industry-wide basis in providing a
collective good would be highly complex, would depend largely on the particular characteristics of the
industry and the value of the collective good, and is beyond the scope of this thesis.
This also bears a strong resemblance to the classic "free-rider" problem, which is, in fact, where the
collective action dilemma finds its roots.
provided in small groups, depending upon the value of the good and the size of the
individuals. The role of group size and structure is basis for further analysis.
The result of Olson's analysis is that, in the absence of other independent incentives
or coercion, the amount an individual actor is willing to contribute to a collective good is
dependent on the fraction of the collective benefits he receives. If the benefits received are
high enough relative to the increase in group costs, an actor will be willing to pay the total
cost for a limited amount of the good.
Group size is one good indicator of the likelihood that individuals will provide
some of the good. Thus in small groups, where the benefits are shared by only a few
actors, it is more likely that some individual player will receive such a significant part of
the collective benefit that he will gain from providing the good, even if he has to pay the
total costs himself.54 In large groups, where each actor only shares in a small part of the
collective benefits, it is likely that only a small amount of the collective good will be
provided.
The size of the 'fractions' and the willingness of individuals to pay for collective
goods can also depend upon the importance of the good and the relative size of players
(i.e., concentration) in the group. If the good is very important either to the group or to
some particular individual then benefits will be much greater than costs and it is more
likely that some individual will find it in his interest to pay for some amount. Similarly, if
there is a concentration of benefits to some actor in the group -- for example, in a mixed
group with both large and small economic players -- it is more likely that some individuals
will be willing to pay for some of the good.
Note that even in small groups the amount of the collective good provided is likely
to be suboptimal. For each dollar put into providing the collective good, the individual
only receives a fraction of the total group benefit; thus, he will discontinue his purchase of
540lson,
34.
the collective good before the optimal amount for the whole group is provided."s The
larger the group, the less of the total benefit will be appropriated by any individual, and the
farther from optimality the group will fall.56
A larger group size also makes it less likely that group pressure will force mcmbers
to contribute. For small or medium-sized groups, it is possible to imagine a variety of
institutional arrangements to provide for collective goods. 57 So long as the group is small
enough that the actions of an individual within the group are noticed by the other group
members, oligopolistic and game theoretic interactions have a chance to affect behavior.
For example, if a group is small and the individual is large, a group agreement to
contribute to the collective good is easily enforced because if the individual backs out,
everyone will know and will themselves stop participating. A rational individual might
continue to pay for the good, knowing that if he stopped everyone else would stop and
there would be no good. On the other hand, if the group is so large that nobody notices
what an individual does or does not do, there is no incentive to contribute.5s Finally, the
larger the group, the higher the transaction costs involved in coordinating group activities.
This will raise the total costs, and create an even greater barrier to group action. 59
The suboptimalities and inefficiency will be somewhat less in a group with
members of greatly different size or interest inthe collective good. In such groups, there
tends to be an unequal sharing of the burden of providing the good. The largest group
5SThis is the problem of a "positive externality," where benefits to other parties are not internalized in the
cost/benefit decisions made by individuals at a microeconomics level. A classic example is the "network
externality" effect often associated with telephone systems or data communications networks. See Katz
and Shapiro, 822-841.
56Olson, 34-35.
"For example, binding treaties among nations or contracts negotiated between parties.
80Olson, 35-36.
59Mansfield, 495.
members, who would tend to provide some of the good on their own, bear a
disproportionate amount of the cost:
Once a smaller member has the amount of the collective good he gets free from the
largest member, he has more than he would have purchased for himself, and has no
incentive to obtain any of the collective good at his own expense. In small groups
with common interests there is accordingly a surprising tendencyfor the
"exploitation"of the great by the small. (emphasis from original) 60
This "exploitation of the great" is a counterintuitive result of the collective action dilemma.
There are other forces which may mitigate the effects of collective action dilemma.
The presence of other incentives at work on the collective group, if they act in a way
similar to the collective good itself, can reinforce the purchase of the good. For example,
in the context of standards, other benefits of supporting the standards development
process, such as access to new technologies or early review of the final standards
produced, can lead individuals to support the process when they might not have done so
just to create standards alone. Coercion can also work to promote a collective good. 61
The net result of Olson's analysis is that small groups are more likely to further
common interests than large ones, and intermediate sized groups may or may not be able
to provide a collective good. These distinctions make it possible to classify groups, and
make predictions based of their behavior.
3.2.3. A Taxonomy of GroLis: Privileged, Intermediate, and Latent Groups
The qualitative differences in behavior among various types of groups make it
possible to create a "taxonomy" of groups based on their size and other characteristics,
60Olson, 35.
example, the loss of access to technology or the inability to influence market technical standards can
be a powerful incentive; influence over standards can be a very effective tool for players with market
power. See Adams and Brock, 36. For example, in an industry such as that of computers comprised of
"an integrated systems seller holding a monopoly in one component market plus a number of specialized
firms producing one or more system components..." the monopolistic seller can manipulate the other
specialized firms "by introducing incompatibilities or undisclosed alterations."
61For
and their ability to act collectively. Olson identifies three different group types, which he
terms "privileged", "intermediate", and "latent":62
Privileged Groups - These privileged groups are the groups referred to above as
small. They are groups composed of few enough individuals that it becomes fairly certain
that some provision of the collective good will be attained. In these privileged groups, the
benefit an individual will accrue will likely outweigh any cost to that individual. It is also
likely that acquiring the good will be beneficial enough to the individual that he will be
willing to contribute his share of the cost. Finally, inthese small groups, it becomes
obvious when an individual is not contributing his share of cost, which is not the case in
large groups. Once that lack of contribution is noticed by other group members, they may
also decide to stop paying, in which case no one will benefit from the common good, and
it is unlikely that a benefiting individual will want to take this risk.
Intermediate Groups - These are groups the size of which is exactly implied by
their name; they are medium-sized, or may contain a mix of both small and large players.
A medium-sized group which may be small enough to make a lack of contribution
noticeable, yet not small enough to provide individual members with some significant
fraction of a collective good may, or may not, be as successful as a privileged group. If an
individual recognizes the reaction that his behavior may cause in others (other members
might stop contributing as well, causing cessation of the common good), he may or may
not decide to continue contributing.
In an intermediate group which has some large and some small members, however,
the resulting cost-to-benefit ratio will likely be skewed. The larger members, who can
more afford the cost, will pay more to attain their share of the common good. Smaller
members may contribute, but will likely stop once they have reached the level of benefit
from the collective good that is sufficient for them. This will be before the larger members
62
01son, 43-52.
have reached a comparable level; as the larger members continue to pay, the smaller
members will be able to benefit from the common good at no cost to themselves.
Latent Groups - Latent groups are those which are characteristically very large and
atomistic. They are likely to have the least success in providing their individual members
with a common good. This is so for several reasons. First of all, the cost of their
organization and coordination is high, which immediately costs each individual more in
order to accrue benefits from the common good. Secondly, it will go unnoticed if an
individual member does not contribute anything to the cost of the common good. Each
individual contributes a very small share of the total cost, and therefore creates no
noticeable difference when she withdraws her contribution. Finally, the benefit accrued to
each individual will not be incentive enough for the member to want to contribute to the
cost of the common good; the benefits will be too thinly spread among the members to
make the costs worth their while.
3.3. Applying Collective Action Theory as a Predictive Model for Standards Setting
As noted in the preceding pages, standards mechanisms are generally a collective
good. They exhibit the characteristics of both nonexcludability and nonrivalry within an
industry, and benefit all users of the standards, regardless of whether a particular user has
paid any of the cost toward the development or maintenance of the standards. It is the
contention of this paper that collective action theory, as defined by Mancur Olson, should
serve as a predictive model of whether or not different sized and organized industries are
able to attain and enact effective standards mechanisms, when such mechanisms are
defined as collective goods.
3.3.1. Predictions
Industry characteristics imply a group classification. Each industry will have
similar stakeholders, in varying numbers. These will include producers, buyers, sellers,
consumers, and often the government and the general public. The number of individuals
involved in an industry will to some extent influence the ability of that industry to
effectively produce industry standards mechanisms.
Insofar as industry-wide standards are concerned, the importance of the sought
standards are crucial when attempting to predict the ability of an industry to develop and
maintain those standards. If it is the perception of those in the industry that such standards
will be economically beneficial, there will be more willingness on the part of individual
members to contribute to the cost of developing those standards. If a lack of such
standards makes competition difficult, or hurts the industry's public image such that it is
negatively affected economically, there will again be more willingness to work on
developing the necessary standards.
Based on the preceding information, the predicted outcomes of the case studies
should be straightforward. If an industry is small in numbers, cohesive and wellorganized, it can be classified as a privileged group, and expected to function accordingly.
It should be successful in attaining the common goods for its members; it should be able to
successfully set accepted and used standards. A medium-sized industry, or one which is
skewed between larger and smaller interested members, it should be difficult to predict
with a high level of accuracy. The importance of establishing the standards would first
need to be known, and even then one of the members may decide that for her the
standards are not so important. In the unbalanced example of an industry, the standards
will probably be attained, but at a greater cost to some than to others within the group. If
an industry is very large and atomistic, it should fail to provide any acceptable level of the
common good for its members. It should be too fractionalized to be able to provide its
members with any incentive to contribute to the cost of achieving the common good;
33.2. Other Factors
Despite the preceding predictions, it should be noted that there will be other
factors at work in determining whether or not industry standards will be set. Groups with
very strong, committed, visible, or rich and powerful personalities working for them may
have some greater ability to attain their goal, regardless of their size. A combination of
strong leadership with wisdom can be a very persuasive tool.
On the other hand, high transaction costs may actually inhibit the procurement of
the common goods in a group that would otherwise be expected to perform well. If the
initial costs of organization and dissemination are overbearing, a lobbying effort may
never get off the ground.
Still other factors make it unclear as to how a group may fare in its attempts to
achieve its goals. If one figures in such things as altruism, community or professional
responsibility, or community or professional standards of conduct, the definitions provided
by Olson may become insufficient to accurately predict group behavior.
3.3.2. Potential Policy Implications
There are several potential policy implications indicated if the theory of collective
action proves correct. First, there is the revelation that groups do not work together to
attain collective goods if the costs to the individuals exceeds any benefit they will accrue.
In industries where this situation exists, there may be a necessity for some sort of third
party intervention. The question then becomes whether or not this third party should be
the government, and if so, to what extent?
Perhaps simply the "threat" of government intervention, or regulation of an
industry, will be enough incentive for an industry to work more effectively toward
developing and setting standards. If that implicit threat is not enough, what should the
next step be? The government may need to appoint an overseeing body to produce and
enforce industry standards. If so, it will need to be determined of whom that body will be
composed, and how it will act in a timely and effective fashion.
International policy implications must also be addressed. The groups forming
standards in the United States will increasingly need to act in accordance with similar
groups worldwide. Perhaps U.S. standards will not be adopted in all cases. Industries will
need to be able to effectively adopt foreign standards in order to stay competitive. It
would behoove industries to keep up to date on foreign developments, in order to be able
to provide the necessary incentives to members so that they will participate in standards
development at a maximum level.
Table 2 - Collective Action Predictions
Industry
Characteristics
Large; atomistic
No one entity
Group Type
Standards
Infrastructure?
Latent
No
Intermediate
Yes
Comments
influences outcomes
Medium-sized;
some large players
or good very
important
Small; large players, Privileged
whose actions are
noticeable
4.
Industry Case Studies
Yes
Exploitation of the
large
4.1. Introduction
As was described in Chapter 3, collective action theory provides one means for
predicting the success or failure of standards mechanisms within a given industry.
Collective action theory provides a model for predicting whether a group will be able to
provide a collective good. In the context of standards, this model can be used to predict
the presence and success of an effective standards infrastructure within a given industry
based upon whether the group is a privileged, intermediate, or latent one.
This section presents case studies of the use, development, and relative success of
standards mechanisms in four different U.S. industries: telecommunications, agriculture,
medical devices, and machine tools. The data for these studies is part of the data from a
study conducted by the U.S. Congress Office of Technology Assessment (OTA) in the
summer of 199163, broadly covering standards-setting processes within the United States.
The cases presented represent each of the types of groups identified by collective action
theory: privileged (telecommunications), latent (agriculture), and intermediate (medical
devices and machine tools). As such, they form a basis for testing the hypothesis spelled
out in Chapter 3: that collective goods theory provides a good model for understanding
the presence of standards infrastructures in individual industries.
4.1.1. Selection of Cases
The cases presented here were selected to meet certain basic characteristics." In
all of them, standards play an important role, both for promoting efficiency, trade and
630TA,
GlobalStandards. See generally.
640ther industry cases considered include aerospace, computers, advanced materials,chemicals, and ship
building.
competitiveness, and from the perspective of a broader national interest in standardssetting.63 All of them have some component of international trade, so industry's
relationship with the international standards community is important as well. For all of
them there is a viable government interest in standards setting, although to varying
degrees. To some extent the availability of data, both published and in the form of
knowledgeable individuals willing to communicate their views, was also a factor. The
group also prevents a spectrum of competitive successes; for example, the
telecommunications industry is a bright spot inthe U.S. economy while the machine tools
sector is clearly struggling.
The cases were also selected to be representative of the group space defined by
collective action theory. At the outset of this study the choice of these industries was
determined based on their size, structure, and market concentration. Cases were chosen to
provide representation of each of Olson's group types: latent (agriculture), privileged
(telecommunications), and intermediate (machine tools and medical devices).
4.1.2. Methodoloeg
The data presented here consist of case studies of different industries, and are to
some extent inherently subjective and non-quantitative in nature. They are intended to
convey both hard data on the industry in question as well as more general "sense" of how
standards work inthe industry. The primary impressions and data presented were gathered
as part of the author's work on the Office of Technology Assessment's 1991 report on
standards for the House Subcommittee of Space, Science, and Technology. Thus, the
questions were asked and answered within the context of Congressional investigative
65For example,
in the medical devices industry, standards are both a basis for trade (as a guarantee of
compatibility or quality) and an important means of regulation (as a guarantee of product safety).
effort. This is noted to provide a context within which to judge the veracity,
predisposition, and motives of the sources interviewed and consulted for these cases.
The primary sources for the cases include library research, reading, and interviews.
The interviews were particularly noteworthy for their depth and subjective portrayal of the
impressions of a broad cross-section of individuals within the standards world. The cases
here are based on dozens of interviews with members of business, government, academia,
think tanks and consumer groups. Included were CEOs of corporations, government
agency officials, "well-knownuniversity professors, and noted consumer advocates. An
attempt was made to meet with people who could provide a broad overview within the
standards process, as well as people who were experts in their particular fields.
Each of the cases conforms to the same rough structure, intended to present both
background information as well as direct analysis of the industry in terms of collective
action:
Description of the industry - Broad background on the industry, including its size
and scope, the relative concentration of firms, its importance within the U.S.
economic picture, the role of global trade, and the relative success and prosperity
of the US industry at this time. Aside from providing the basis for classifying the
industry according to Olson's group size and concentration paradigm, the
background is intended to provide a more colorful and textured context against
which to judge and understand what is happening.
The role of standards - An overview of how standards are used within the industry,
and their significance. The importance of standards plays a role in classifying the
type of group within the collective action model. It is also intended to shed light on
the role of standards more generally in society, as well as specifically in the given
industry.
Analysis: the success of standards and collective action - A measure of the relative
success of standards efforts, providing a somewhat subjective assessment of the
level of investment in standards infrastructure, the ability of those efforts to meet
users needs, and the success of U.S. standards organizations both nationally and
globally. This assessment is then considered in the context of the predictions of
collective action theory. Finally, other factors of possible influence besides
collective action are offered for consideration.
4.1.3. Predictions
The collective action model described in Chapter 3 provides the basis for
predictions on the level of investment in standards infrastructure to be expected in each of
the cases. For the latent case, agriculture, the model predicts a relatively low level of
organization in standards-setting, correlating to underinvestment in standards. For the
privileged groups, a relatively sophisticated standards-setting infrastructure is predicted,
with significant investment by individual players in the industry. For the intermediate
cases, the outcome is indeterminate.
For a summary.of these predictions, see Table 2 - Case Study Predictions.
Industry
Table 3 - Case Study Predictions
Standards
Group Type
Infrastructure?
Agriculture
Latent
Medical Devices
Intermediate
Disorganized,
underinvestment
Indeterminate
Machine Tools
Intermediate
Indeterminate
Telecommunications
Privileged
Highly organized
Comments
Might expect
exploitation of large
Might expect
exploitation of large
4.2. Agriculture
USDA Grade Standards for US Agricultural Products
United States Department of Agriculture (USDA) standards for agricultural
products are a ubiquitous part of the American consumer's life. From "USDA Choice"
beef to "Grade A" eggs, USDA grades are a source of consumer confidence and an
important measure for merchants of the quality and value of products. The USDA takes
an active role in grading dairy products, fruits and vegetables, meat, poultry, shell eggs,
livestock, cotton, tobacco, and wool. This grading process in turn rests on a series of
nationally uniform standards developed and issued by the USDA and used throughout the
world.
The agriculture industry -- farmers, processors, wholesalers, and retailers -- is an
example of a latent group. The industry is highly atomistic, with many small and mediumsized actors and few large players. Until the first half of this century the development of
grade standards was left in the hands of private industry. This mechanism failed to produce
badly needed standards in a timely fashion. Today, almost all agriculture product standards
are set by the USDA. The failure of the market to provide this collective good -- and the
evolution to the current government-sponsored standards process -- are the subject of this
case study.
4.2.1. Aericulture in America
Agriculture in America is likely as close to perfect competition as any US industry.
The industry encompasses the production and distribution of produce, meat, eggs,
tobacco, cotton, wool, and other products.6 Highly atomistic in character, the producers
66U.S. Department of Agriculture, Agricultural Marketing Service, This isAMS (Washington, D.C.:
GPO, 1991), 3.
- farmers - have traditionally been small in size and large in numbers. They are
geographically diverse and there is virtually no major concentration of market power
among producers, wholesalers, or retailers. The output of any individual player has no
significant impact on the price of goods. Even today, as the average size of individual
farms has risen and the number of farms has dropped, the market remains highly
segmented and competitive.
The United States began as an agrarian society, with the bulk of agricultural sales
in the local communities where they were produced. With the increase of industrialization
in the nation through the nineteenth century, the proportion of farm products not
consumed locally but sold in urban markets steadily increased. 67 Several factors
contributed to the growing national agriculture market. Improved transportation was
crucial, especially for perishable goods like fruits and vegetables. Advances inthe financial
markets allowed for a cash based market, whereas before goods had largely been
exchanged. The national market proved more efficient as farmers found increasing returns
from specialization. A budding communications infrastructure allowed for control of this
market.
By the turn of the century, agriculture inthe US had ceased to be a local industry.
Some products remained regionally specific, while others like cotton, tobacco, and flour
were already national inscope. With these changes came complications. There were
conflicts between buyer and seller over the measures of quantity and quality. Where before
trade had been based on individual lots open to inspection by the purchasers, the national
market meant that buyers had to order goods sight unseen. There was a need for basic,
agreed, standardized measures of the quantity and quality of goods.
67 Sharon
Bomer, Assistant to the Director, Fruit and Vegetable Division, interview with author,
Washington, D.C., 22 July 1991.
4.2.2. Standards and A'ricultural Products
"There was a time," notes the lead character in the Broadway musical Fiddleron
the Roof, "when he sold him a horse, but delivered a mule." The story of agricultural
grading standards is a story of defining the quality of goods as a basis for trade. As the
size of the average American farm grew through the late nineteenth century, buyers and
sellers of agricultural products were increasingly hard-pressed to agree upon - and verify the quality of goods. (This was especially true for fruits and vegetables, where quality
from a given farm could fluctuate widely depending upon ripeness, shipping conditions,
and seasonal variations.) Without agreements on quality, it was hard to set prices or
resolve contract disputes.
Standards served several purposes. Standards were crucial to buyers in carrying on
trade over long distances. The price of goods meant little when a buyer could not verify
the quality, uniformity, or even quantity of produce inthe bushels being shipped.68
Producers quickly realized there was a premium to be made from higher quality
products. 69 More specialized areas began to market higher quality goods with regional
names, such as Genessee flour or Goshen butter (areas in western New York). Over time,
goods of similar quality from other regions were sold with the regional name, and a
standard was created.
"6Bomer, interview with author:
"The problem was a lack of communication or, more precisely, a lack of understanding in
whatever communication was taking place. A seller on the West Coast could honestly describe
his product to a buyer on the East Coast and the buyer could develop and entirely inaccurate idea
of what he was purchasing. An attempt to defraud may not even have been involved. The
problem was a simple matter of lack of agreement on a trading language."
See also U.S. Department of Agriculture, "Core Speech: Fresh Products Branch," nd., TS (photocopy
of typescript), 2.
69
Clarnce H.Danhof, Change in Agriculture: The Northern UnitedStates, 1820-1874mbndge,
MA: Harvard University Press, 1969), 43. "Itmade no difference to a farm family M•potatoes were
an assortment of white and red varieties, but such mixtures suffered significant discounts when offered
for sale."
Standards in Use - Box 1
USDA Cotton Standards
Cotton is a product where precise product standards have had a direct impact on
competition and efficiency. New automated textile manufacturing processes rely on ever
greater knowledge about the properties of the materials to be used. Being able to insure
that cotton lots meet tight tolerances over a range of factors such as texture, fiber length,
color, and seediness allows for increasingly efficient new processes. Being able to
accurately grade produce also provides a more precise indication of their quality and
value.
To facilitate this process, the USDA has developed a series of 39 grade standards
and 15 staple lengths for Upland cotton (the native U.S. species). Each year the USDA
distributes descriptive standards as well as a cottons standards "kit" to growers and
buyers, containing dozens of physical samples for comparison of color, texture, graininess,
seediness and other characteristics over a range of grades.
In additions, the USDA runs seven inspection facilities across the country for
official grading of cotton. (While inspection is voluntary, upwards of 98% of all cotton
produced in the U.S. is inspected -- in part because inspection is required for participation
in certain government-sponsored growers programs.) Every bale of cotton is inspected
and graded independently, for a total of 16 million bales per year. This level of resolution
matched in few other countries -- for example, the Soviet Union only inspects lots of 15 to
16 bales. The inspection process in the U.S. has become increasingly automated, with
complex new technologies enabling inspectors to grade a greater level of refinement with
greater efficiency.
The U.S.'s leadership in inspection technology has made it a leader in global
standards as well, with USDA cotton standards and grading techniques adopted widely as
international standards. The sophistication of the U.S. standards and inspection process,
and the ability to precisely grade each bale of cotton produced, has improved the value of
the cotton, increased the efficiency of manufacturing processes, and yielded real
competitive advantage for U.S. cotton growers and textile manufacturers.
Sources: United States DepartmentofAgriculture (SharonBomer, FredMullins)
Where standards could be adopted, they were successful. One example was the
Chicago grain trade:
For the first time anywhere, it forced into the grain trade huge quantities of
uniform warehouse receipts based on a definite quantity and grade rather than an
individual lot in store....Grades were first indicated on the receipts at Chicago in
1857. The practice spread to other markets and became the keystone of a
spectacular, worldwide, market structure.70
Agreeing upon standards required a degree of coordination and market control that did
not exist in most of the disparate agricultural markets, and product grades were slow to
develop. As the market mechanism grappled for ways to provides this collective good,
farmers and food buyers turned to third party graders to set standards and inspect goods
as they went to market.71
4.2.3. Evolution of the USDA's Role
The early 1900s saw the rise of regional third-party graders who enforced local de
facto standards for food quality. Third party grading introduced its own problems, and a
set of emerging tensions contributed to the eventual breakdown of private-sector grading
systems:
1.
There was significant contention between producers, buyers, and graders over who
should pay for grading and how standards should be set. De facto regional
standards tended to give grading companies a local monopoly, and there was little
recourse in appeals.
2.
The lack of unified standards between grading companies still made price setting
difficult. As the developing transportation infrastructure turned the market for
agricultural goods into a national one, inconsistencies between regional standards
became more difficult to reconcile.
3.
As U.S. crop exports increased, merchants realized that guaranteeing a minimum
standard of quality for US products was vital for ensuring their reputation in the
70Agricultural historian G.A. Lee, as cited in James R. Beniger, The ControlRevolution: Technology and
the Economic Originsofthe Information Society (Cambridge, MA: Harvard University Press, 1986),
165.
71Fred Mullins,
Cotton Division, interview with author, Washington, D.C., 22 July 1991.
international marketplace. 72 It came to be felt that there was a need for a
centralized standard setter.73
4.
Finally, in the wake of Upton Sinclair's The Jungle and government regulation of
the meat-packing industry, the Federal government was inclined to take a more
active interest in facilitating the grading of agricultural goods.
As the third-party market mechanism proved unable to meet the demands of the
market, the government began to take a more active role in product standards, often at the
behest of industry. Through a series of legislative acts, such as the U.S. Apple Grading
Law of 1912 or the Cotton Standards Act, the Congress provided established mechanisms
for setting standards.74 The passage of the Agricultural Marketing Act of 1946 gave the
USDA the broad authority it exercises today in setting grading standards and establishing
inspection facilities for a wide variety of agricultural products."
As mentioned above, issues of public health and safety, as typified in the aftermath
of Upton Sinclair's The Jungle, were already prompting the government to take a more
active role in the regulation of food and drugs. In addition, the exigencies of the First
World War provided an impetus for government grading and inspection of produce for use
in the war effort. Once established, these efforts became important to the industry and
remained after the war. 76 These two factors combined to influence the USDA's
involvement.
include "Florida oranges"and "Idaho potatoes.".
7Mullins, interview with author.
72Examples
74U.S.D.A.,
This isAMS, 24-26. For example, the Apple Grading Law specified dimension for the
standard apple barrel, provided certain requirements for standard grades of apples, and set penalties for
misuse of government standards.
See alsoU.S.D.A., "Core Speech," 3.
75 Kenneth C.Clayton, Deputy Administrator, letter to author, Washington, D.C., 10 July 1991. See also
U.S.D.A., This isAMS, 24.
76 U.S.D.A., "Core Speech," 3.
4.2.4. Agricultural Product Standards Today
Today the USDA's Agricultural Marketing Service(AMS) is the primary grader,
inspector, and standard setter for American agricultural products. AMS grades meat,
poultry, eggs, dairy plants, fruit, vegetables, cotton, tobacco, and other products. Some
grade marks are widely seen by consumers; others are largely for the benefit of food
wholesalers or processors, and may not be visible at the retail level. All grade marks are
determined by AMS standards "based on practical attributes that determine the value and
utility of the product."" The AMS has developed a complex and sophisticated set of over
400 standards78 that vary widely from product to product. For example, there are eight
USDA beef grades, but over 155 grades for some varieties of tobacco. 79 The development
and adoption of such a sophisticated set of standards has in turn made it possible for
manufacturers and processors to use more efficient, streamlined, and sophisticated
methods, improving efficiency and competitiveness.8 0 Use of a USDA standard symbol
requires inspection by a certified federal or federal-state inspector.8 •
Compliance with USDA standards is completely voluntary; only in the case of
certain safety regulations is inspection of goods required by law. It should be noted,
however, that large government food purchases which require USDA grades -- such as
military procurements or Free Lunch programs -- can make government standards
7U.S.D.A., This isAMS, 8-9. "The standards for the quality of beef, for instance, describe such factors
as the amount of marbling (fat interspersed within the lean), the color, firmness, texture, and age of the
animal for each of the grades. These factors are indicators of the tenderness, juiciness, and flavor of the
meat - the characteristics of beef that are important to consumers."
" 8Clayton, letter to author.
79U.S.D.A., This isAMS, 8.
soBomer, interview with author. Mullins, interview with author.
slU.S.D.A., This isAMS, 9. Inspection is a large and important part of AMS's mission and its ability to
enforce uniform standards. As noted in Chapter 2, the issues surrounding inspection services are
complex and in many ways separate from the establishment of standards. As such they are beyond the
scope of this thesis.
virtually mandatory. 82 The USDA standards process is financed by the producers of
goods, who pay fees for inspections. Inspection fees are a small part of final costs -typically 1-2% of the goods. AMS claims that its inspection and grading operations are
run on a break-even basis.93
Standards also play an important role today in the global market for agricultural
products." In a highly competitive and often protectionist world agricultural market
place, standards can have considerable impact on the U.S. producer's ability to trade.
Inconsistent standards between countries make it difficult to trade for known quantity and
quality of goods. Standards have also been used as a "non-tariff barrier to trade" by
countries seeking to remove competition from exports. s5 In addition, there may be
significant economic benefits to be derived fi ,nthe adoption of U.S. national standards
by the international community.86
For all of these reasons international standards are becoming increasingly
important and receive attention at high levels of government. The USDA represents the
industry in the Codex Alimentarius Commission, the primary world food standards
82Bomer,
interview with author.
83Bomer,
interview with author. Mullins, interview with ai4hor. The U.S.D.A. even claims that a 1985
fee increase met with positive response from industry, who saw it as a way to further improve the
inspection process.
4The United States exports over $2billion worth of agricultural products annually.
"8Bomer, interview with author. For example, in the late 1980s the European Common Market used a
new definition for its high grade of avocado that required an oil content just below that of U.S. avocados
- thereby shutting out all U.S. suppliers. Safety has also been used as an excuse to prohibit imports. In
1988, U.S. beef exports to Europe were halted because of the presence of steroids deemed "unhealthful"
by the Europeans; the FDA considered these substances safe, and it was widely believed that the
standards were an excuse to curb U.S. beef exports.
86Mullins,
interview with the author. For example, local producer will have greater knowledge of what it
take to meet a national standard. It is easier to be certified to a national standard, rather than having to
have foreign inspectors certify to a foreign standard. Finally, greater producer involvement in the
standard process tends to shape those standards to the producers' advantage.
organization, as well as in other international standards organizations such as the ISO or
the Universal Cotton Agreement. 87
Where private-sector efforts to provide standards had proven insufficient by the
turn of the century, the USDA standards mechanisms have been a success inmany
important ways:
1.
The USDA has become the repository for much of the industry's expertise
concerning standards. Unlike other industries where the firms possess the bulk of
knowledge about a particular standard, in many cases it is the graders and standard
setters working for the USDA who actually have the greatest understanding of the
needs of farmers and food buyers.88 In part this is a function of the size and
technical sophistication of the average food buyer or seller. It may also be
attributed to the increasing complexity of inspection techniques. 89
2.
The USDA has gotten high marks for developing grading techniques and
technology. In many cases, US standards and inspection methods are among the
most advanced in the world, and this has contributed to the successful adoption of
U.S. standards as the de facto world standards 90 [See box on Cotton]
3.
The USDA has been largely successful in having U.S. standards adopted
internationally. This can likely be attributed to the market power of the US in the
global agricultural market, as well as the sophistication of its standards. However,
it is only through the coordinated exercise of this market power that the US has
been successful.
97U.S. Department of Agriculture, Agricultural Marketing Service, AgriculturalMarketing Service
InternationalPrograms(Washington, D.C.: GPO, 1991), 8-10.
88Bomer, interview with author.
"Mullins, interview with author.
"0Mullins, interview with author.
4.
The USDA standard-setting process is funded through user fees on inspections,
and is thus largely financially self-sufficient.
5.
The government's active role in standard setting has allowed public interests such
as food safety to be addressed directly. It also removes any threat of privatesector monopolies over standards.
4.2.5. Summary and Analysis: Agriculture and Collective Action
The mechanism for developing agricultural standards is a collective good. For
agricultural producers, processors, and sellers the existence of standards is crucial to
trade, and an important factor ininternational competitiveness. At the same time, the
development and maintenance of standards is costly ano requires a high degree of
coordination.
At the start of this century, and to a large extent today, the farmers and sellers with
a stake inthe development of standards constitut(ed)a "latent group." The market was
highly atomistic, with many small actors and little coordination. There were few actors, if
any, whose participation or lack of participation in the standards-setting process would be
widely influential. There were no players large enough that the development of standards
was so important that it was worth funding a national standard-setting effort.
This latent group proved unable to provide itself with sufficient quantities of the
collective good. De facto standards introduced were successful only in limited situations.91
The markets efforts to foster third-party regional standard setters broke down due to the
conflicts over who would pay for the services and how they would be coordinated.
Despite the acknowledged value of the collective good, the group was unable to set
standards in a meaningful way and was forced to resort to government intervention.
91For certain
products, like wheat, where infact there was a greater degree of market concentration. Inthe
case of wheat, the Chicago mercantile was able to exercise considerable market authority and imposed
its standards on the entire wheat industry.
Aside from collective goods theory, there may have been other factors that
influenced the evolution to the USDA standards mechanism.
1. Regulatory interests - There was a government interest in public health and safety,
although this played out primarily in the effort of the Food and Drug
Administration.
2. World War I - The needs of the federal government procurement agents during the
war played a role in the USDA's initial involvement.
3. International trade - International treaties on food safety standards, representation
ininternational standards bodies, and the effective exercise of U.S. market power
on the global scene require a high degree of central coordination. 92
4. Inspection - The tight coupling of the USDA's standards setting with its huge
inspection operation gives the government tighter control on the use of its
standards, provides a source of revenue for standards operations, and makes the
USDA the primary source of expertise instandards issues and technologies.
From a prescriptive, policy-making perspective, the success of the USDA
standards setting efforts provides some important lessons about the possibilities of
government intervention in standards. The USDA success at providing timely,
technologically sophisticated standards, in a cost-effective manner, with the accumulation
of considerable expertise within the ranks of the government, flies in the face of
conventional wisdom about what government can do. While some would argue that the
USDA standards and inspection operations have swelled the ranks of the Washington
bureaucracy, the success of the USDA may indicate that public institutions are viable
solutions to the standards collective action dilemma insome cases.
2However, in other industries such as telecommunications the requirements of treaties and international
trade negotiations are similar and the government plays no special role in standards setting. For
example, in telecommunications the State Department and Trade Representative work directly with
industry standards organizations indeveloping national positions.
4.3. Medical Devices
From heart valve specifications to breast implant regulations, medical device
standards play an important role in assuring safety and promoting trade. Standards are a
critical mark of quality used by buyers and sellers to acknowledge the value of devices.
Moreover, in the absence of other mechanisms standards have become a de facto means of
regulation by the government. The cost of complying with different national standards has
made them an important consideration for international trade. Finally, they serve to assure
device users of their health and safety.
The medical device industry is an example of an intermediate group, with both
small firms and large multi-national corporations who maintain significant market share.
By many measures, medical devices represent one of the success stories of U.S. standard
setting. A well-organized US standards effort -- marked by high levels of industry
participation 93, active government cooperation 94, and industry-wide coordination of
standard setting groups", and an effective presence in the international standards
community -- has grown in response to threats of regulation, the rise of international
trade, and the felt needs of the professional health care community. At the same time, the
US standards process faces continual challenges from consumer groups who contend that
the current standards process does not adequately ensure the safety of medical devices
users. The apparent success of US medical device standards is also threatened by rapidly
changing technology and the rise of regionalization in the global marketplace.
93Michael
Miller, Executive Director, Association for the Advancement of Medical
Instrumentation, interview with author, Washington, D.C., 14 July 1992. Sharon Stanford,
American Dental Association, interview with author, Washington, D.C., 13 July 1992.
94According to the FDA, the agency is well-represented at most standards meetings. The current
chairman of ANSI's Medical Devices Standards Board is from the FDA. Trade organizations
concur with this assessment of the FDA's role. Miller and Stanford, interviews with authors.
95ANSI's Medical
Devices Standards Board serves as a meeting ground for the many different
standards setting organizations. The Board has over forty members, meets quarterly, and is
generally well-attended. Steve Cornish, telephone interview with OTA Staff, 28 March 1991.
4.3.1. The Medical Devices Industry
Medical devices are a striking feature of modern health care, encompassing X-ray
and electromedical equipment, surgical and medical instruments, surgical appliances and
supplies, dental equipment, and opthamalic goods. They have improved the quality of
lives and revolutionized the practice of medicine. Some devices, such as the artificial hip,
help to improve an individual's functioning after an otherwise debilitating injury; others,
such as the kidney dialysis machine or pacemaker, serve to extend lives. Medical devices
have also dramatically altered medical diagnosis and treatment. Blood chemistry analyzers
have made sophisticated laboratory tests available to all, while new imaging devices have
replaced dangerous exploratory surgery. Finally, more common devices such as contact
lenses or dental products are a ubiquitous part of American life today.96
The manufacture and sales of medical devices in the United States is a multibillion
dollar industry characterized by rapid growth, a high degree of innovation, competitive
success internationally, and great diversity inboth products and firms.97 From less than $1
billion in 1958, industry sales have grown to over $17 billion in 1983. 98 Research and
innovation has been the hallmark of this growth, with devices that did not exist a decade
ago now a commonplace part of the market. The U.S. industry is a world leader
technologically, and has maintained a strong competitive position in the global market.99
9U.S. Congress, Office of Technology Assessment, FederalPoliciesand the MedicalDevices Industry
(Washington, D.C.: GPO, 1984), 3.
"70TA, FederalPoliciesandthe MedicalDevices Industry, 17.
"This represents more than a six-fold increase in real dollars. OTA, FederalPoliciesandthe Medical
DevicesIndustry, 19. The OTA assessment also notes some of the difficulties in quantifying the size and
scope of the industry. The most comprehensive statistics come from the Department of Commerce's
Census ofManufacturers, which categorizes products by Standard Industrial Classification (SIC) codes.
"Mike Miller, Executive Director, Association for the Advancement of Medical Instrumentation,
telephone interview with author, 13 August 1991.
The U.S. has run a growing trade surplus in medical devices through the 1970s and 1980s,
in sharp contrast to the total U.S. merchandise trade deficits throughout this period.'10 In
terms of growth in companies, employment, new products, and foreign trade, the medical
devices industry has exceeded manufacturing as a whole as well as similar manufacturing
sectors.'0 '
The structure of the medical devices industry includes numerous small companies
as well as large ones with significant market power. Small companies are responsible for a
greater percentage of sales than is typical of manufacturing industries, and small firms
have played a particularly important role in the introduction of new products.102 However,
large, often multinational, corporations -- such as Johnson & Johnson, Baxter-Travenol,
and American Hospital Supplies -- account for the majority of sales, and there is a degree
of market concentration.103 A small number of firms control a considerable share of the
market, particularly within specific product lines.'"
The nature of medical devices attaches especial importance to their safety and
quality, and has introduced a wide variety of stakeholders with interest in the workings of
the industry. In addition to the manufacturers, interested parties include hospitals (who
account for the majority of device sales), health professionals, medical patients, and
individual consumers. The broad public interest in device safety -- and the importance of
technological expertise in assessing safety -- has led to a strong government regulatory
interest in medical devices.
100lbid. Also,
OTA, FederalPoliciesand the Medical Devices Industry, 35. The U.S. has run a trade
deficit in merchandising inall but two years (1975 and 1976) since 1973.
01 0OTA,
1021bid.,
FederalPoliciesand the Medical Devices Industry, 37.
17, 37.
10 3Ibid., 25-27, 37.
"•T~hefour leading firms account for approximately 32 to 45 percent of sales in each medical device SIC
code, as compared to 43 percent average in US manufacturing. OTA, FederalPoliciesandthe Medical
Devices Industry, 28-30, 37.
Regulation of US Medical Devices
The Federal Food, Drug, and Cosmetic Act of 1938 first authorized government
regulation of medical devices. The Act gave the Food and Drug Administration (FDA)
authority to stop the sale of adulterated, misbranded, or unsafe devices, but only after they
were on the market. The burden of proof fell on the government to show that devices on
the market were in fact unsafe. Through the late 1960s, it became apparent that the
regulatory framework was insufficient to guarantee the health and safety of device users.
A notable 1969 report by the Department of Health, Education, and Welfare estimated
that over a 10-year period, 10,000 injuries and 731 deaths were associated with medical
devices.105 Faulty artificial heart valves contributed to 512 of these deaths, and this wellpublicized problem -- coupled with other problems including faulty defibrillator circuitry,
overheating infant incubators, and obstructed tracheotomy tubes -- prompted
Congressional action in 1976.'06
The Medical Device Amendments of 1976 gave the FDA much broader powers to
insure device safety and efficacy. Under the act pre-market approval is required for some
devices, and all device makers must register with the FDA. Devices are regulated on the
basis of their potential risk. The act places a device into one of three risk categories:' 07
Class I - General Controls: Devices with the lowest potential risk to patient health, for
which "general controls" are reasonable assurance of safety. Tongue depressors and
contact lenses are examples of Class 1 devices. The general controls imposed include FDA
notification and conformance to "good manufacturing practices" applying to the
manufacture, storage, shipping, and installation of devices.
1050TA,
FederalPoliciesand the Medical Devices Industry, 97-98.
06Sidney Wolfe, The Public Citizen, telephone interview with author, 10 July 1991. See also OTA,
FederalPoliciesand the Medical Devices Industry, 98.
FederalPoliciesand the Medical DevicesIndustry, 98-99, 109.
1070TA,
Class II - Performance Standards: Devices for which general controls are considered
insufficient, such as x-ray machines and anesthesia equipment. Devices in this category are
to meet mandatory performance standards designed to insure their safety and efficacy.
Class III - Pre-market Notification: Devices which support life, prevent impairment, or
present unreasonable risks of injury. Examples include pacemakers, heart valves, and
contraceptive intrauterine devices (IUD). These devices must pass strict pre-market
approval processes, and have their safety and efficacy documented.
The FDA, while slow to initially promulgate its regulations, 08s has established
procedures for Class I device control, including good manufacturing practices for all
device classes. It has also promulgated and is enforcing pre-market approval requirements
for Class III devices.109 The bulk of devices, however, have been classified in Class II,
which has become the de facto catch all category. These devices are to be regulated
through performance standards, but to date the FDA has only been able to formally
develop standards for none of the more than 1200 Class II device types."11 The standards
development process has proven far more time-consuming than initially anticipated; the
FDA has adopted a policy of reliance on manufacturer conformance to voluntary
consensus standards developed inthe private sector
4.3.2. Standards and Medical Devices
Product standards for medical devices provide guidelines for the design,
manufacture, and use of devices. They are used by industry, health care facilities, medical
professionals, and the government to dictate the design, procurement, and regulation of
10
o
By 1984, for example, the FDA had only succeeded in classifying 11 of 19 device groups. Ibid., 105.
1°91bid.,
99-110.
"lbid.,109. Wolfe, telephone interview. Gene Kimmelman, interview with author, Washington, D.C.,
22 July 1991. The FDA has not formally adopted any Class II standards to date.
these devices. These product standards are designed to achieve two overriding goals:
quality and safety.
Quality standards certify that products meet expectations about performance,
reliability, or durability; for example, the American Dental Society's standards for dental
fillings serve to guarantee fit, stability, and sterility. [See Box 2] The primary value of
these standards is often as a marketing tool by producers, and a guarantee for buyers of
the quality of goods. They may also serve as compatibility standards to provide for agreed
upon ways of using devices or the integration of devices, and to improve or simplify the
use of devices.
It is safety standards which have received the most attention in the medical devices
community. These standards govern the manufacture and use of devices to maximize
patient safety. They are often highly technical in nature, and the complexity and criticality
of products make their adoption especially important. For example, a typical health-care
professional purchasing X-ray equipment is unlikely to have a full understanding of the
electrical and mechanical design features necessary to ensure proper product functioning;
rather, he will rely on industry standards and certification as an indicator that products are
actually safe.
Standards in the Real World - Box 2
American Dental Association Standards
The American Dental Association (ADA), a professional organization of dentists,
device and supply manufacturers, technicians, and other health care providers, is the main
proponent of standards for dental products in the United States. For products ranging
from toothbrushes and dental floss to sophisticated x-ray equipment and dental implants,
the ADA develops and propagates standards by coordinating ANSI activities, participating
in international bodies, and administering the ADA Seal certifying product compliance.
ADA standards are used as an important measure of quality. For example,
standards for dental amalgam (used in fillings) govern such technical characteristics as
"creep compliance," "dimensional stability," and the presence of foreign matter. These
standards are then used by dentists -- and the public -- to assist in buying decisions. The
importance of complying with ADA standards is not lost on the industry. As one Proctor
& Gamble Co. executive has noted, "Ifyou look at dentifrice in the past two decades,
about 27 new ones entered the market without the ADA seal and not one is still on the
market."
Given the public health ramifications of its products, the ADA's standards have an
important safety dimension as well. While many of the products covered by ADA's
standards are subject to FDA Class II mandatory standards, in the absence of any FDA
standards the ADA standards have become the de facto regulations for safety. Since the
ADA is primarily a professional organization serving dentists, the safety aspect of its
standards are given high regard. However, consumer safety groups have noted that they
have no actual representation on any ADA standards committees.
By many accounts, the ADA's efforts are successful. They have been successful at
developing standards in over 45 different product areas, and maintain an active presence in
the ISO Dental Technical Committee. ADA officials cite to three primary reasons for their
success: the professional ethos of the medical profession, the potential threat of imposed
FDA regulations, and the contributions of individual and organizations who staff and
support the standards process. Sources estimate that volunteers contribute over $2 million
per year in donated services, as compared with less than $100,000 spent by the ADA itself
in the actual development of standards.
Sources: Sharon Stanford, American DentalAssociation; Journal of the American Dental
Association, April 1987.
Medical device standards in the United States have traditionally been developed by
private-sector trade associations, professional organizations, and industry groups. These
standards are non-binding'"I, and are established by consensus organizations made up of
11IClass
I standards are approved by the FDA and are mandatory.
various industry participants, including doctors, nurses, manufacturers, and hospital
administrators. The bulk of medical device standards today are approved by ANSI, whose
Medical Devices Standards Board oversees processes for developing device standards and
receives active participation from industry and government. Much of the actual standards
development work is done through large trade organizations, such as the American
Association for the Advancement of Medical Instrumentation(AAMI), the Health Industry
Manufacturer's Association(HIMA), or the American Dental Association (ADA); some
technical standards are developed by specific technical organizations such as the
Association for Standards and Testing of Materials(ASTM) or the Institute of Electrical
and Electronic Engineers(IEEE). The development work largely conforms to ANSI's
consensus standards development practices, and are open to those with the resources and
desire to participate. In the end though, "standards succeed if they are adopted."
Internationally, the bulk of global medical device standards are developed through
the International Standards Organization (ISO).12 Other international standards are
developed in relevant organizations such as the International Electrotechnical Committee
(IEC). The U.S. is represented inthese bodies through the American member body,
ANSI. U.S. companies play an active role both on U.S. TAGs and in leadership positions
in ISO and IEC meetings, technical committees, and working groups. Of equal concern for
American device makers, however, is the rise of regional standards organizations,
particularly within the European Community (EC). The EC intends to use control medical
device safety through standards developed by CEN and CENELEC, organizations to
which US companies are not entitled representation. At worst, it is felt that these EC
standards are ina position to impose non-tariffbarriers to trade on U.S. device makers; at
112 Stanford,
letter to author. Miller, telephone interview with author. The ISO maintains a number of
technical committees which include work on medical devices and instrumentation, such as ISO/TC1066
for Dentistry, or ISO/TC 198, for Sterilization of health care products.
the least, they are a serious competitive disadvantage for U.S. sellers in the European
market. 113
4.3.3, Summary and Analysis: Medical Devices and Collective Action
The U.S. medical devices industry seems to represent a success story in American
standards setting. There is a highly developed, and well-supported, standards
infrastructure for the creation and propagation of standards. The ANSI Medical Devices
Standards Board serves as a clearinghouse for standardization issues, and receives wide
support from industry, individual health care providers, hospitals, and the government.
Major professional and trade organizations, such as AAMI, HIMA, and the ADA, provide
extensive support, both in dollars and manpower, to the various national and international
standards efforts. 114 These efforts are paid for through the dues assessed to organization
members, and through the voluntary contributions by stakeholders of technical resources,
money, and manpower. The major trade organizations do not finance the standards efforts
by sales of the standards. These organizations have formally recognized the importance of
standards setting activities, and for many it is viewed as their most important service to
their constituent members. All of these efforts receive broad-based industry support: from
small companies, who understand their importance to effective trade; from big companies,
who also understand the implications for international trade; from health care providers,
who use standards to ensure quality and safety; and from government, which increasingly
uses private-sector standards as a means of regulation. The government has been a
3Miller, telephone interview with author. Miller, letter to author. "The EC plans to use standards as ...a
primary means for obtaining the EC mark, essential to the sale of products in all countries of Western
Europe."
114Standards development comes from the dues paid by member organization, companies, and
individuals. Dues are assessed on a sliding scale based on the member's ability to pay. Development is
also paid for by the sale of standards, although to a much smaller degree than other industries.
particularly strong supporter of the process, with FDA providing technical and personnel
resources, including the current chairman of the ANSI Medical Devices Standards Board
who is an FDA official.I15
The success of the process has manifested itself in several ways, not the least of
which is the timely development of a broad array of standards for devices. Both
manufacturers and product users and buyers perceive the process as a relatively effective
one that meets their needs for topical, agreed upon, readily available standards developed
in short time periods and covering the functional areas needed.116 Standards are also
perceived as crucial marketing tools. Finally, standards are considered to meet the health
and safety needs of the profession and the public at large -- as indicated by the fact that
there have been few efforts to discard the process, and the government considers it a
sufficient form of regulation.11"7
The U.S. also enjoys success on the international standards front. Within the ISO,
American companies and organizations support much of the work being done inthe
working groups, and now enjoy positions of responsibility and authority within the various
working groups and committees of the ISO. The U.S. has been inordinately successful in
adopting American standards as ISO standards. This can be attributed in part to the
industry's position of technical leadership, its market share (both as a global producer and
as an important consumer), and it's leadership in standards development itself. The
industry's success is also attributable to the highly skilled and sophisticated individuals
who represent the U.S. to the international standards community. They are well-trained,
115 Steve Cornish, Letter to OTA Staff, 9 July 1991.
"16 Miller, interview with author. Stanford, telephone interview with author. Bernard Liebler, interview
with author, 10 July 1992. Dr. Robert Duestenberg, Deputy Assistant Secretary of Commerce, and
Charles Ludoff, interview with OTA staff, Washington, D.C., 25 July 1991.
117James
J. McCue, Jr., FDA, telephone interview with author, 5 August 1992. The FDA has felt
comfortable enough with private standards to leave all Class II device standards setting in the hands of
the consensus standards process. Moreover, an in depth study of seven private standards has found that
they provide as much safety as those developed specifically thorough government.
conversant in the language, customs, and methods of the international business
community, have many years of experience, and are highly respected within the
international community.
The U.S. has also achieved a measure of success even within the closed standards
setting processes of the European Community." •8 The U.S. standards representatives
abroad have been very successful at getting CEN and CENELEC to adopt many of the
ISO standards for the EC, and have achieved ISO harmonization with CEN/CENELEC
standards where appropriate. Even within the closed-door EC process, U.S. interests have
been represented by large multi-national corporations such as Johnson and Johnson, who
are allowed to participate because of their status within the parent countries. The MNCs
have worked closely with their American counterparts and have been an important source
of information for standards developers and trade negotiators about the inner workings of
the EC standards process.
The importance of the standards infrastructure is evidenced by the industry's
growing strategic approach to standards development. As Mike Miller, Executive Director
of AAMI noted, "AAMI has secured influential national and international leadership
These leadership
positions essential to the development of important standards ...
positions is ISO and IEC enable the United States to maintain contact with national,
regional, and international standards development and have close access to channels of
communication and influence that determine the content of national, regional, and
international standards."'
19
This position appears typical of a sentiment common
throughout the industry -- including manufacturers, users, and regulators -- of the
11 Duestenberg,
interview with OTA Staff. Miller, interview with author. For example, in the Fall of
1991 U.S. standards representatives were able to achieve a self-described 'coup' by getting the European
Community to agree to adopt ISO standards - rather than develop their own - for several board
categories of devices. The US medical device standards community has developed a comprehensive
strategy for dealing with EC 1992, which includes efforts at harmonization, influencing EC standards
thorough MNCs, and pressuring the EC to open up its pFocess.
S19Miller, letter to author.
acknowledged strategic importance of mechanisms for the development of device
standards.120
Several factors play a role in the success standards infrastructure found in the
medical devices industry. 121 Underlying all of these is the industry's willingness to support
these activities thorough money and manpower, and its recognition of the strategic
importance of standards. Other issues playing a role include:
The threat of regulation - Industry participants note that it was the threat of
1.
heavy-handed FDA regulation of Class II devices that mobilized the medical
devices community in the late 1970s to invest heavily in standards. In effect, the
industry sought to head off FDA standards by showing it could do the job itself.
The effort worked, and the FDA has felt increasingly comfortable with entrusting
Class II regulation to private standards. Moreover, the FDA's threats served as
"the perfect dress rehearsal for EC 1992."122 The industry's efforts to "get its
house in order in the early 1980s"' 23 ended up placing it in a particularly good
position to deal with the regional standards efforts of CEN and CENELEC.
Government partnership - The FDA has been an active participant in standards
2.
efforts, providing direct input into the process. There has also been only one, clear
government agency, as compared to the disparate agencies to be dealt with in
other industries.
Clear standards goals - The goals of device standards are relatively clear -- i.e.,
3.
to promote safety and quality -- as compared with other industries such as personal
computers or high definition television where the "right answers" are not so clear.
20Miller, interview with author. Liebler, interview with author. Duestenberg, interview with OTA
Staff..
121See
generally, Miller, Stanford, Cornish, Duestenberg, McCue.
telephone interview with author.
122Miller,
123Ibid.
4.
Industry prosperity and leadership - The U.S. industry has been both profitable
and dominant, allowing it the luxury of spending money on standards and being
effective when it does so. There is indication of a 'rich get richer' syndrome here,
where those who do well are able to strategically invest in standards, and thus do
better still.
5.
Payment structure - The costs of development is born largely by corporations
and health care providers, either through dues to trade organizations or by direct
support of standards committees. This makes the standards setters directly
accountable to the standards users.124
6.
Personal leadership - The industry has been well served by the personal efforts of
several high profile leaders who have been extraordinarily effective at getting U.S.
standards adopted abroad and promoting investment in standards at home.
The one major failing of the medical devices standards process has been its
inability to include direct representation of consumers within the standards setting process.
While corporations, health professionals, and their representatives are active participants
in the standards infrastructure, the various standards bodies do not include any regular
presence by consumers or consumer watchdog groups such as The Public Citizen or
Consumer Federation of America.125
Consumer advocates believe that the absence of consumers leads to standards that
are unsafe and do not meet the needs of the general public. They cite continuing device
failures and problems as evidence of neglect. While the government does take an active
124 Contrast
this with other systems, where standards are supported by the sale of standards. This has the
unfortunate effect of placing the incentive on selling more standards, not necessarily on creating the best
standard. For example, the ASME boiler code is hundred of pages long, a best seller, and a huge source
of revenue for ASME. However, the EC is unwilling to adopt such a voluminous standard. While it is in
ASME's members' best interests to adopt a more compact standard that can be harmonized with the EC,
ASME has thus far been unwilling to compromise for fear of losing the income from its old standard.
125 Kimmelman,
interview with author.
role in standards setting -- ostensibly as the representative of the people -- the consumer
groups consider many of the official involved to be part of the problem, co-opted by the
industry they work so closely with. Consumer advocates also consider the shift from
mandatory, FDA-developed performance standards for Class H devices to private-sector
standards to be an abdication of responsibility on the part of the FDA. The FDA, for its
part, maintains that the private standards do in fact provide the requisite level of safety
required.
26
The reasons cited for the lack of consumer involvement stem largely from the costs
of participation inthe standards process. Consumer groups largely do not have the
personnel with the expertise and resources available to industry members who sit on
standards bodies. It is estimated that supporting a consumer group representative would
cost approximately $50,000 per standard developed.127 The dispersed nature of the
benefits from such participation -- spread among millions of device users in the general
public -- make it virtually impossible for consumer groups to finance effective standards
body representation.128
An FDA study in the mid-1980s reviewed seven different private-sector standards for Class II devices,
for which data was readily available, and concluded that the private standards did provide adequately for
the health and safety of the public even in the absence of direct government involvement. McCue,
telephone interview with author.
126
1 27
Miller, telephone interview with author.
128 The
groups instead opt for more cost-effective ways of achieving their ends, such as litigation. From
their perspective, litigation offers a cheap way to achieve large recoveries and influence corporate
behavior in the quest to limit liability. Of course, from a broad social welfare point of view it is probably
more efficient to have consumer interests represented up front, in the standards process, rather than
through lawsuits after people get hurt. Kimmelman, interview with author.
4.4. Machine Tools
Standards in the machine tools industry are an important means of coordination
and a source of real competitive advantage. Tool standards are used to guarantee
compatibility, insure safety and quality, and integrate control of tools into the
manufacturing process. A successful national standards infrastructure can greatly enhance
the usefulness and capabilities of tools. Effective representation internationally can have
strategic value. This value in standards is being played out against the backdrop of a
troubled industry facing stiff competition and a difficult contraction.
The machine tools industry today is an intermediate group; it is composed of many
small companies and a few large ones. Despite the importance of standards, the industry
has been largely unable to establish an effective standards infrastructure. Particularly in the
international standards arena, the U.S. has had difficulty getting positions adopted and has
found itself playing catch up with foreign standards organizations which use standards as a
strategic marketing tool. The industry also exhibits "exploitation of the large," with bigger
companies bearing most of the cost of standard-setting and smaller companies paying very
little.
4.4.1. Machine Tools in America
The machine tools industry in America is marked by its vital importance to
manufacturing and its steady decline in the United States over the last two decades.
Machine tools are "power driven machines, not portable by hand, used to shape or form
metal by cutting, impact, pressure, electrical techniques, or a combination of these
processes."' 29 The machine tools industry is critical to America's manufacturing base:
It is not possible to have world-class manufacturing without world-class
tools. It is not possible to have world-class tools if one has to import
129 "Machine Tools:
1981), 2.
America's Muscles," (McLean, VA: National Machine Tool Builders Association,
them all. Yet the American machine tool industry is dissolving. In 1964
the United States was a net exporter; in 1986, 50 percent of its machine
tools were imported. Production is now only half of what it was at the
peak. 130
US manufacturers have begun to rely on foreign tool makers for their machine tools.
equipment. Foreign vendors provide quality tools, with relatively quick delivery, at a
good price. But this dependence has dire consequences for US manufacturers: chances
are good that state-of-the art tools purchased abroad have already been in use in that
country for some time.' 3 1
Through the early 1970s the United States dominated the machine tools industry.
U.S. firms enjoyed a world-wide reputation for technical excellence, quality, and price
competitiveness. It is an insular industry; many American tools makers are small, familyowned businesses, filling a "niche" in the local or national market. Many were established
to fill a particular manufacturing need, mainly in localized areas, and saw no need to
expand or innovate so long as the manufacturing base remained constant and profitable.' 32
Japanese and European tool makers made inroads into the market in the 1970s.
American producers were ill-equipped to deal with the competition for a variety of
reasons: under investment in new tools and manufacturing technologies; unwise business
practices, and long times to delivery; a lag in standardization of new numericallycontrolled (NC) tools; a dramatic under-estimation of the threat of competition.
Moreover, consumer demand seems to have shrunk permanently in the U.S.; process
improvements have greatly reduced the need for machining. 133 When recession hit in
13°Michael L. Dertouzos, et al., Made in America: Regainingthe ProductiveEdge (MIT Press, 1989) 20.
13 1"The director
of a recent GM study on machine tools laments:
If you by the very best from Japan, it has already been in Toyota Motors for two years, and if you
buy from West Germany, it has already been with BMW for a year-and-a-half."
"The U.S. Machine Tool Industry and Its Foreign Competitors," Working Papersof the MIT Commission
on IndustrialProductivity(Cambridge, MA: The MIT Press, 1989), 3. (hereafter, MIT Working Papers).
13 2
33
John Alic,
Office of Technology Assessment, interview with author, Washington, DC, 8 July 1991.
Drtouzos, et al., Made in America, 235.
1981, the highly cyclical industry took a nose-dive, many orders went to cheaper foreign
competition, and the American tool industry was savaged. Total U.S. shipments dropped
from $5.1 billion in 1981 to $2.1 billion in 1983, and only rebounded to $2.8 billion during
the recovery of 1986-7.134
With the contraction of the early 1980s, many firms went out of business or
merged with larger firms. Today the industry would best be classified as an intermediate
group. There are many small (fewer than 20 employees), family-owned businesses, which
continue to fill specialized niches, as well as several large players such as Cross &Trecker,
Cincinnati Milacron, and Ingersoll Milling.' 35 The industry has begun to stabilize inthe
early 1990s, albeit at a much reduced level and with a dose of new humility.136
4.4.2. Standards and Machine Tools
Standards serve several important functions for the US machine tools industry.
Due to the interaction of the tools, compatibility standards are often critical. Pallets must
meet forklift specifications; die sets must fit tool fixtures. Tools need to be able to
interact with each other, and they also must be able to produce finished products to agreed
upon measurements. The more wide-spread the standards, the more useful the tool. The
inability of many small United States tool makers to widely standardize, inpart due to the
fractious and niche-oriented nature of the industry, was an early source of competitive
losses to the larger, more organized Japanese and German producers. 13 7
13 4M1T
Working Papers, 85. It is estimated that half of all US tool makers went out of business or were
bought out between 1980 and 1990.
135MIT Working Papers, 9, 90. Sixty-five percent of domestic builders have 20 or fewer employees.
136Alic,
interview with author. Anthony Bratkovich, National Machine Tool Builders Association,
interview with author, McLean VA, 10 July 1991. Through the 1980s, industry problems were often
linked to factors such as the high wage rate in the U.S., or the presence of unions. Today, the tenor of
the evaluation ismuch more focused on new strategic innovations and improving efficiency.
137Made
in America, 236-247. Alic, interview with author.
Standards in the Real World - Box 3
Sample Power Press Safety Standard
What does a standard look like? Here is one example.
The ANSI safety standards for mechanical power presses require point-ofoperation devices which stop the tool if the operator's hands are inadvertently place in the
point of operation. This requirement for presses with two-hand controls is a sample of the
type of language found in these standards:
"The safety distance (Ds) between each two-hand control device and the point of
operation shall be greater than the distance determined by the following formula:
Ds - 63 inches/second X Ts
where:
Ds = minimum safety distance (inches);
63 inches/second = hand-speed constant; and
crankshaft rotation (seconds)"
Source: "AmericanNationalStandardforMachine Tools - MechanicalPower PressesSafety Requirementfor Construction,Care, and Use, "ANSI B11.1-1982, American
NationalStandardsInstitute, Inc., 1982, p. 38-40.
Machine tool standards serve an important function as guarantees of product
safety and quality. For buyers, tool compliance with standards is a guarantee of safety to
an often uneducated consumer. For the tool maker, compliance with safety standards is an
important marketing tool as well as a shield against liability. For the government,
standards are sometimes used as a way for OSHA to regulate workplace hazards.' 38
The control of tools has become one of the newest and most important
battlegrounds in standardization. New numerically controlled (NC) machines are able to
hold tolerances more effectively. 13 9 Moreover, the use of computers allows many tools to
138Charles
Carlsson, NMTBA, interview with author, McLean, VA, 10 July 1992. Mr. Carlsson oversees
NMTBA's efforts to develop safety standards.
139Dertouzos, et al., Made
in America, 232.
be controlled from a common source and be more tightly integrated. The United States
was slow to adapt numerically controlled tools and develop a standardized control
language for their integration, resulting in competitive losses through the 1980s.'14
The mechanisms for developing industry standards are scattered, and not clearly
defined. There is no single acknowledged, cooperative group which sponsors standards
research and recommendations industry-wide. ANSI certifies many machine tool
standards, as it does for other industries. Other strong, independent representative
groups, such as ASTM, the American Society of Mechanical Engineering (ASME), and
the American Petroleum Institute (API) develop standards in particular areas.
Participation and support from companies is spotty. The high costs of individual
participation have prompted many smaller companies to turn to trade associations like the
National Machine Tool Builders Association (NMTBA) to represent them. However, the
NMTBA itself does not have sufficient resources to underwrite a first-rate standards
infrastructure. 141
Industry standards efforts have been described as "hampered" by the size, diversity,
and perspective of most tool makers. Serving specialized, local niche market, many tool
makers were late to understand the importance of national standards. The cost of
participation in standards efforts is too high for many. 142 As one NMTBA official noted, a
small die shop cannot afford to send one of its top engineers off to a standards meeting for
days at a time. 143 Often the burden of support for standards efforts fall to the larger
14oM1T Working Papers, 8, 15-17.
141Infact,
the NMTBA receives very little money specifically for standards. Most of the support for
standards activities is revenue from NMTBA's annual sponsorship of the Machine Tool Builders Trade
Show. Carlsson, Bratkovich, NMTBA, interviews with authors.
142 NMTBA serves, to some extent, to combat the lumpiness of costs by distributing them among a large
number of small players. However, as NMTBA's revenue stems largely from other sources, there is little
relation between NMTBA activities and cost sharing. See footnote 9.
143Bratkovich,
interview with author.
companies, who have a larger stake in standards and greater resources to promote their
development. 144
International machine tool standards are set through the ISO and the IEC. The
prominence of many national standards, particularly those of Germany's DIN and the
Japanese, make the sale of tools abroad a complicated affair. Moreover, the European
Community's efforts at harmonization may be a mixed blessing for U.S. tool makers; they
will tend to simplify the field, but can be a subtle, powerful barrier to trade.
The U.S. tool makers all but ignored the international standards community until
1982.145 America's dominance in the market allowed producers to design their tools with
impunity. There was little interest in collaborating to develop and lobby for adoption of
U.S. positions on standards issues. Even as U.S. producers began to see the need for
involvement, their efforts were hampered by the lack of effective mechanisms for
coordination, especially among small manufacturers. Representatives to ISO meetings
lacked the experience and connections of their foreign counterparts, and were often less
than effective in the "old-boys" network of the international standards community. 146
The costs of participation in the international standards process are high, and the
factors that led to failures inthe national process play out similarly for representation
abroad. The only companies who could afford to send personnel to standards meetings
overseas were the large companies, who ended up bearing the brunt of American effort.
As an executive at one large tool maker complained, "It feels like we're being used." 147
The costs of incompatibility with international standards ca• be high. When
American tools were the best in the world, foreign customers were willing to buy non144Bratkovich, interview with author.
145Alic, interview with author. Carlsson, interview
with author. Bratkovich, interview with author. .
146Stories
are told of back-slapping American delegates, ignorant of protocols and etiquette, who alienated
their European hosts with excessive familiarity, bawdy jokes, and well-publicized internal feuding. They
also wouldn't speak anything but English. Bratkovich, interview with author.
147Cincinnati Milacron, telephone interview with author, 22 July 1991.
standardized products. But in the current competitive market, compliance with local
standards can be critical -- and costly, if those standards are not harmonized with
America's. Moreover, some nations see influence over machine tool standards as a
powerful strategic marketing tool for their manufacturing technologies. Germany, for
example, has sent teams of standards experts and spent millions of dollars to facilitate
Saudi Arabia's adoption of many DIN standards. 148 As one NMTBA official noted, "Ifyou
need a tool, and you have the DIN catalog sitting on your desk, who are you going to
call?"149
4.4.3. Summary and Analysis: Machine Tools and Collective Action
The success of the United States machine tools industry in its use of collective
action to coordinate standards development has been minimal, at best. It has not been able
to support a strong, effective standards infrastructure despite the value of such a process
to all members of the group and the clear competitive losses stemming from its absence.
In the context of collective action, the industry is an intermediate group, with a fair
number of small players and some large ones as well. For intermediate groups the ability
to provide a collective good is indeterminate. In this case, the industry structure and
mechanisms made it unlikely that the support or absence of support of any but a few large
players would be noticed by the group as a whole. Moreover, the importance of the good
to individual players, and particularly theperceived importance, was low enough and the
1480TA,
Global Standards, 36. Also, Dr. Robert Duestenberg, Assistant Secretary of Commerce,
interview with OTA staff, Washington, D.C.
149Bratkovich, interview with author. Bratkovich also noted that Russia's GOST standards "are
practically all DIN standards," and DIN has also made inroads in China. The Saudi Arabia case
received some notoriety in America, prompting Congress to mandate aDepartment of Commerce
response. Eventually, a retired U.S. standards official was sent, but the mission had to be privately
funded and lasted less than a year (in contrast to the mutli-year, five-person, government sponsored DIN
effort.) OTA, GlobalStandards, 36.
costs of support high enough that it was not worthwhile to any but a few to contribute
regardless of what others did.
The industry has also experienced the "exploitation of the large" posited by
Olson. 15o Costs for developing industry standards thus far have been mostly born by the
larger companies. There is a feeling among these larger corporations that they are unfairly
bearing the lion's share of the costs, and being used by the smaller companies.151 Smaller
companies seem satisfied to adopt the standards which result from the process, but not to
take part in the process or bear a share of the cost burden. However, they run the risk of
alienating their benefactors.
In addition to the inability of the US machine tools industry to employ the tenets of
collective action to attain its collective goals, there are other factors that may play a role in
the development of industry standards processes:
1. Industry profitab-iity - An industry mired in recession, lay-offs, and competitive
losses has few resources to devote to standards efforts, particularly where the competitive
pay-offs may not come for years.
2. Understanding and education - There is little understanding on the part of the United
States machine tools producers of the importance of standardization, particularly within
the new global economy.
3. Personalities and animosities - As noted above, the inexperience and lack of savvy of
members of the industry standards community may have played a role in the lack of
success of the efforts to develop national standards and influence international standards.
Moreover, well known personal and organizational animosities, such as the conflicts
between ASTM and ANSI, have been detrimental.
150
Olson, 35.
151Cicimnati
ilacron, telephone interview with author.
3. Little threat of regulation - Aside from safety standards, the federal government has
exhibited little or no intention of regulating the industry. Thus the threat of regulation has
had little effect. The government has had some impact, however, in de facto standard
setting as a large customer.
4. Metric measurement - The fact that the United States still functions on the English
system of measurement is a major stumbling block toward international competitiveness.
4.5. Telecommunications
Committee TI-Telecommunications Standards
Telecommunications standards are the life-blood of the industry. The utility of the
communications network depends upon its ability to reach as many users as possible. In a
far-flung and ubiquitous network this ability to interact relies, at a very basic level, on
widely accepted standards. These standards are particularly important given the rapid
growth of telecommunications and its crucial role in the emerging information economy.
In the context of collective action, the telecommunications industry is a privileged
group. It consists of a relatively small number of stakeholders, each of which has a strong,
vested interest in the success of the standards process. The standards infrastructure is
highly organized and well supported through large investments of money, time, and
personnel devoted to the research and implementation of industry standards. Yet despite
this highly developed standards process, questions still remain about the capabilities of the
standards infrastructure and its ability to incorporate a wide variety of interests,
particularly consumers.
4.5.1. Telecommunications in America
Telecommunications in America is a multi-billion dollar industry characterized by
rapid technological change, a strong position in the global marketplace, and growing
importance within the national economy. In less than 150 years the industry has grown
from the simple telegraph to the brink of a complex digital, multi-media, optical network.
Today the information needs of society are receiving greater and greater attention, having
been elevated so far as to become a rallying cry in a successful presidential campaign.' 5 2
152"To create
millions of high-wage jobs and smooth our transition from a defense to a peacetime
economy, we will rebuild America and develop the world's best communication, transportation, and
environmental systems. .... We will focus on four critical areas: ... [2] A national information
network to link every home, business, lab, classroom, and library by the year 2015. To expand access to
The telecommunications industry finds itself a critical element in the nexus between
information, computer and communications systems considered crucial to success in the
emerging global economy.'5 3
Through the early 1980s the U.S. telecommunications industry was marked by the
dominance of AT&T. Functioning as a regulated monopoly, AT&T exercised control
over nearly every facet of the telecommunications infrastructure in America. Its worldwide
technological leadership made it a dominant player globally as well. The 1982 consent
decree that broke up the Bell system introduced competition into the industry. While
removing sole responsibility for integration of the phone system from one organization, it
introduced new players and opened new markets. 154
Today the number of companies involved inthe industry has grown, although it
remains relatively small. The major players are large, often multi-national, corporations,
and include: exchange carriers, such as the Bell Atlantic or US WEST; interexchange
carriers, such as AT&T, MCI, and US Sprint; manufacturers, such as Motorola, Northern
Telecom, Alcatel, IBM, and DEC; and, increasingly, other media companies such as
Time-Warner or NBC. The new faces and players make coordination that much more
difficult, and the need for effective coordination mechanisms, standards, that much more
critical. 155
information, we will put public records, databases, libraries, and educational materials on line for public
use." Bill Clinton and Albert Gore, Putting People First:How We Can All ChangeAmerica (New
York: Times Books, 1992) 9-10.
153 Drake, William J.and Lee McKnight, "Telecommunications Standards in the Global Information
Economy: The Impact of Deregulation and Commercialization," in ProjectPromethee Perspectives
(March 1988).
t s4 For a history of telecommunications in America and the recent impacts of the AT&T break-up, see for
example: Richard Solomon, "What Happened After Bell Spilled the Acid?", Telecommunications
Policy, June 1978; U.S. Congress Office of Technology Assessment, CriticalConnections:
Communicationsfor the Future (Washington, D.C.: GPO, 1990).
IssDrake and McKnight, Promethee (March 1988).
4.5.2. Standards and Telecommunications
By specifying the technical features of components within the network, standards
serve to either facilitate or hinder the development of wide-spread, robust communications
systems. These technical standards range from physical characteristics such as modular
jack dimensions or wiring voltages, to sophisticated standards regarding communications
protocols or switching software, to process standards such as how to bill customers and
what a busy signal means.
Standards in telecommunications serve to promote compatibility and
interoperability; reduce variety to achieve economies of scale and scope; provide
information about complex goods; and guarantee quality through minimum levels of
performance.' 56 Standards also have a role to play insuring the integrity of the
communications network. Agreed upon protocols and design specifications can be used
to prevent devices from interfering with the network, particularly in case of
malfunction. 157
The standards picture in telecommunications is clouded by overlap with other
technologies and industries. The "convergence of modes" brought on by digital
technologies, as posited by Ithiel De Sola Pool, has blurred the lines between the
traditional telephone system and fields like data communications, information technology,
and entertainment. Standards that were once the exclusive province of telephone
156
Marvin Sirbu, as cited in Lee McKnight, "The International Standardization of Telecommunications
Services and Equipment," in E.J. Mestmaecker, ed., The Law andEconomics oflnternational
Telecommunications (Baden-Baden: Nomos Verlag, 1987) 419.
157 This issue represents
the FCC's initial foray into the world of communications standards, beginning
with the Carterfone in the late 1960s, and remains a large part of the FCC's perceived role in standards
today. Julius Knapp, Jr., Office of Engineering and Technology, Federal Communications Commission,
interview with author, Washington, D.C., 8 July 1991.
companies are now receiving attention from computer makers, cable operators, movie
studios, and newspapers.158
Prior to the break-up of the Bell system, telecommunications standards were
established by AT&T on a defacto basis. These standards were published as the Bell
Practices. The practices were an internal tool for AT&T's use, but were widely
distributed outside of the company as well. Given the dominance of the American market
and AT&T's technological leadership, Bell Practices were often adopted internationally."59
Following the break-up of AT&T, the need for coordination was quickly
recognized by the regional Bell companies and other players in the market. These new
stakeholders collaborated, through the Exchange Carriers Standards Association (ECSA)
and with the blessings of the FCC, to create Standards Committee T1Telecommunications in 1983. The Committee was accredited by ANSI, and is now the
primary consensus standards body responsible for developing U.S. telecommunications
standards.' 60
Committee TI functions as a voluntary consensus standards organization. Its
myriad committees, subcommittees, and working groups bring together engineers,
managers and other experts from industry, academia, and the government to formulate
complex technical specifications as well as U.S. positions for international standards.
Committee T1 is committed to an "open door" policy; any member of the committee,
including foreign members, may attend any meeting and play a role in the consensus
1s8 Ithiel
de Sola Pool, Technologies ofFreedom (Cambridge, MA: Belknap Press, 1983). For a telling
recent example, consider Congressman Edward Markey's admonition to the computer industry to get
involved in the telecommunications standards debate: "Decisions will be made this year.that will affect
the rest of [the computer industry's] lives, and unless you participate, the rules will be cut in a way that
will not be to your advantage." Steve Lohr, "Computer Makers Told to Get Involved in Rules," New
York Times, February 23, 1993, D2.
159Philip Johnson, AT&T Bell Laboratories, interview with author, San Francisco, CA. 6 August 1991.
16olan Lifchus, "Standards Committee TI-Telecommunications," IEEE Communications Magazine,
January 1985, 34-37. Committee TI-Telecommunications, Annual Report, 1990, 1, 26-28.
adoption process.16 ' Committee T1 is supported by member dues, but the real support for
the standards process comes from the donation of personnel and other resources to the
meetings and mailings that are the basis of the standards development process. 162
Other standards committees also play a part intelecommunications standards.
ANSI Committee X3, sponsored by the Computer Business Equipment Manufacturers
Association (CBEMA) oversees many computer standards. The Institute of Electronic
and Electrical Engineers (IEEE) administers standards committees, most notably in the
area of data communications. Other organizations, like the Electronics Industry
Association or ASTM, are concerned with physical components and other characteristics
of the phone system.
In the United States, the federal government plays a limited role in the
development and implementation of standards. The two organizations associated most
with telecommunications standards are the Federal Communications Commission(FCC)
and the National Institute of Standards and Technology(NIST). Relevant Congressional
subcommittees also weigh in occasionally. All of these agencies, however, serve a minimal
role in standards development on a regular basis. If involved at all, it is more likely for
particularly visible standards, on an ad-hoc basis. 163
161TI's
open door policy has been the source of much controversy. Some cotwmentators feel that TI's
openness, incontrast with the closed strategy sessions of its foreign counterparts, amounts to giving
away the keys to the candy shop. Richard Solomon, telephone interview with author, 27 June 1991. The
Committee's leadership disagrees. Ivor Knight, TI Chairman, argues that the open exchange of ideas is
the Committee's greatest strength, especially inaworld of unclear national boundaries. Even though
companies occasionally lose the advantage of acquiring certain standards information ahead of others
Knight feels that the advantages of openness far outweigh the losses from secrecy. Ivor Knight,
interview with author, Washington, D.C., 8 July 1991.
162 Committee Ti-Telecommunications,Annual Report, 1990, 22-25; Ivor Knight, interview with author.
163FCC interviews. NIST interviews.
The FCC sees itself as a "watchdog", there to ensure the integrity of the
network.164 The FCC sets some standards, mostly regarding restrictions on devices
connecting to the phone system, and occasionally involves itself in larger standards issues
(such as the adoption of ISDN.) On a regular basis, however, the FCC does not
participate in T1 in a substantive way. 65s For example, at a recent TI Executive
Committee meeting there was only one FCC representative, whose role was largely one of
observer; there was no sense that the FCC was attempting to play any sort of leadership
role, or serve as a representative of the public interest.66
NIST views its role in standards development processes even more narrowly. It
sees the Government as a large telecommunications user, and itself as a representative of
that large customer. Its job is to ensure that standards are maintained so that its customer
can function efficiently. This is in contrast with the oft-perceived NIST role as leader of
standards efforts in America.' 6 7
Internationally, telecommunications standards are developed and implemented by
the International Telecommunications Union (ITU) and its component organizations, the
International Consultative Committee for Radio (CCIR) and the International Consultative
Committee for Telegraph and Telephone (CCITT). The ITU is a treaty organization, a
United Nations specialized agency, and the Unites States delegation is led by the State
Department.'68 As of the Winter of 1993 the ITU is in the process of restructuring,
164Knapp, interview with author.
165
Knight, interview with author.
166Committee TI-Telecommunications Advisory Committee Meeting, San Francisco, CA, 5-6 August
1991.
167James Burroughs,
NIST, interview with OTA staff, Gaithersburg, MD, 29 July 1991. Mr. Burroughs,
who oversaw NISTs mandate of the GOSIP data communications protocol, noted that NIST is actively
interested in the standards process as a user, but its interest does not come from any overarching
directive to improve communications standards for the nation.
168McKnight,
418-419.
combining the functions of the CCIR and the CCITT into one, as yet unnamed, standards
directorate.
The State Department supervises the selection of the U.S. delegation to ITU
meetings. However, the State Department does not see its role as that of active
representative of the public interest. The Department's representatives tend to have little
technical industry experience, and perceive themselves as serving a "process" role, rather
than a pro-active one in actual development and implementation. The State Department
attempts to act in ITU proceedings to reflect the industry's position on standards, not the
Government's. '69
The international standards picture is further complicated by the presence of
regional organizations, such as the European Telecomrmunications Standards Institute
(ETSI), as well as national groups, like AFNOR in France and DIN in Germany. The
convergence of communications and computing has also led to overlap with other nongovernmental standards organizations such as the ISO (which developed the important
data communications Open Systems Interconnect reference model), the IEC, and the new
joint ISO-IEC venture Joint Technical Committee-i (JTC-1). Furthermore, the growing
role of the Internet Engineering Task Force(IETF) as a focal point for internetworking
standards is arguably, in many ways, overtaking the importance of the ITU and ISO in
these areas. With U.S. involvement from TI, X3, ECSA, CBEMA, IEEE, and ANSI;
federal government agencies including the FCC, NIST, and the State Department; and
international organizations including ITU, ISO, IEC, JTC-1, ETSI, as well as the national
groups, the international standards picture is exceedingly complex.
4.5.3. Summary and Analysis: Telecommunications and Colective Action
'6 9U.S. Department of State, interview with OTA staff Washington, D.C., 12 June 1991.
The mechanism for developing telecommunications industry standards is a
collective good. The committees and organizations that create standards benefit
communication companies, and individual consumers, by providing standards that promote
interoperability and allow for the incorporation of new technologies into the existing
system. To a large extent this benefit accrues to all who use the standards; in a world of
open systems it is difficult to exclude anyone from the use of a standard.
Within the context of collective action, the telecommunications industry
constitutes a privileged group. The size of the group is relatively small; small enough, at
least, that non-participation in a collective effort by any single member is likely to be
noticed. Moreover, the collective good is of great importance to each member of the
group. Without standards, the industry could not function.
At least interms of investment in standards, the telecommunications industry is a
success story. The group has maintained a vast, complex, expensive infrastructure of
committees, subcommittees, and working groups to provide needed standards. Individual
organizations consistently invest time, money, and personnel in a group effort which relies
largely on the expertise of valued individuals. The industry has been able to sustain this
infrastructure without any regulation, coercion, or punitive sanction against those who do
not participate.' 70
Interviews with industry participants in Committee Ti reflect a belief that, at least
within the U.S., the standards infrastructure has been able to meet the complex and diverse
needs of the telecommunications system. The organization also gets high marks for the
quality of its standards, which make great use of experts in the field. Finally, there also
appears to be an understanding on the part of industry of the need for, and advantages of,
support for collective standards mechanisms. 171
170
Knight, interview with author.
171Michael
Varnassi, MCI, interview with author, San Francisco, CA, 6 August 1991. W.J. McNamara,
III, BellSouth Services, interview with author, San Francisco, CA, 5 August 1991. Arthur Reilly, Vice
The industry is not without its problems. The absence of forceful leadership that
comes with authority has a price. Effectiveness on the international scene may be a
casualty of the open, fractious nature of the democratic standards process."72
Development of new standards can be frustratingly slow.173 The process is vulnerable to
manipulation by special interest groups. Finally, some constituencies, such as consumers,
are rarely represented in the process at all. 174
However, it is not clear that any outside agent would be able to replicate the
collective efforts of the industry. The myriad working groups and esoteric standards
generated by Committee Ti provide some indication of the daunting task awaiting any
central planner deciding which standards to develop. The technical expertise, and the
familiarity with the market, required good indications of why a grass-roots, private sector
driven standards mechanism would likely have great advantages over any government or
third-party effort.
In addition to the collective action mechanism at work, there are other factors
which may have contributed to the industry's success in developing standards mechanisms:
1.
Ability to influence outcomes - Companies and individuals receive the added
incentive of influencing the outcome of the standards process by being actively
Chairman, Committee TI-Telecommunications, interview with author, San Francisco, CA, 5 August
1991. Johnson, AT&T Bell Laboratories, interview with author. This view has been supported by other
commentators, including Burroughs, NIST, interview with OTA staff; Oliver Smoot, Vice-President,
CBEMA, interview with author, Washington, D.C., 12 August 1991.
1 2The international standards community is often frustrated with mixed signals from the U.S. In
addition, disagreements within the U.S. standards community are often well-known to the its
competitors and used to its disadvantage. OTA, Global Standards, 12-14.
173This has been one of the major criticisms of the U.S. process. In contrast, the establishment of ETSI in
Europe was specifically designed to accelerate standards development. OTA, Global Standards,70-71.
Knight, interview with author.
interview with author. The expense of participation means that it isnot worth their while
to be part of the standards process. Consumer groups strapped for resources get far more "bang for their
174Kimmelman,
buck" through litigation and lobbying. Unfortunately, in general the cost of dealing with litigation after
the fact is far higher than the cost of incorporating design concerns at an early point in the standards
process.
involved in the process. To the extent individual actors have a stake in which
standards are adopted and are able to impact the process, the standards
mechanisms become, in a sense, collective goods with additional private benefits.
2.
Early access to standards, technology - Participating companies can learn ahead
of time the likelihood of one standard being adopted over another, and prepare
themselves accordingly. They may also benefit to from early access to technology
and strategy from other participants in the process.
3.
Historical roots and professional ethos - The industry has its roots in the
monopolistic days of AT&T, when everyone worked together to build a phone
system for the public. Today the ethos of cooperation, and of the public trust
granted to the industry to provide this vital service to the nation, may be an
important influence in the desire and ability of the industry to organize itself.
5.
Enlightened leadership - The standards community has benefited greatly from
savvy, respected, politically astute committee leaders, as well as corporate leaders
who understand the importance of the investment in standards.
6.
The ITU as treaty organization - The special treaty nature of international
standards may serve as an added incentive to participation.
7.
Threat of regulation - The implicit threat to the industry of regulation by outside
forces, such as the FCC, has put the industry on notice that it must organize itself
or have rules imposed upon it.
8.
Technological leadership - Historically, the U.S. standards effort has benefited
internationally from U.S. technical leadership. While the U.S. is not the dominant
force it once was, technological advances by U.S. industry may still play a role in
the success of the standards process.
4.6. Summary of Results
The data in this study indicates that collective action theory is consistent with the
development of standards mechanisms, processes, and 'infrastructure' within US industries.
(See Table below.) This in turn can have important implications for the policy debate
about standards setting currently taking place within industry and government. This policy
debate, and the implications of the results here, will be discussed in greater detail in
Chapter 5.
Table 4 - Case Study Results
Industry
Group Type
Standards
Comments
Infrastructure?
Agriculture
Latent
Unable to organize
USDA now sets all
Medical Devices
Intermediate
Highly successful
Strategic investment
in standards; threat
Machine Tools
Intermediate
Underinvestment;
Unsuccessful, esp.
in international
Late recognition of
importance of stds;
exploitation of the
arena
large
Very successful;
high levels of
investment
Highly organized;
still concern over
"best" standards
industry standards
of FDA regulation
Telecommunications
Privileged
5. Conclusion and Policy Analysis
There are matters in which the interference of law is required, not to overrule the
judgments of individuals respecting their own interest, but to give effect to that
judgment: they being unable to give effect to it except by concert, which concert
again cannot be effectual unless it receives validity and sanction from the law.
1 75
-- John Stuart Mills, Principles
The national stake in standards, coupled with the apparent failure of the current
system to invest properly in standards mechanisms, presents an important policy problem.
Collective goods analysis and the case studies reviewed in the preceding chapters indicate
that in many cases market mechanisms will not be sufficient to provide appropriate
standards. This in turn has clear ramifications for setting both public and private policy.
This chapter presents an analysis of the policy issues raised and the implications of
the collective goods analysis. First, the nature of the policy problem is defined. Then the
current policy environment, and a range of possible alternatives, is presented. The primary
elements of the policy debate are discussed. Implications and general findings of the
collective action analysis and case studies are then presented, along with their impacts on
the current policy debate. Finally, directions for further research and investigation are
outlined briefly.
5.1. The Policy Problem
Standards are important, and it is not clear that we set them well inAmerica. High
visibility cases such as HDTV, the problems of the mac•"iine tool industry, orthe grappling
175John
Stuart Mills, Principles,Book V,Ch. XI, § 12 as cited in Mark Blaug, Economic Theory in
Retrospect (Cambridge: Cambridge University Press, 1985), 218.
over data communications standards provide anecdotal evidence of what seem to be
breakdowns in the system."76 Further research reveals more fundamental flaws: lack of
coordination between players, particularly within the government; a systemic disregard for
many small players, such as consumers; animosity between the major standards
organizations; and consistent strategic undervaluation of standards and their importance.
This, broadly stated, is the standards policy problem.
The problem has many components. Are the "best" standards being created? Are
they being created quickly enough? Are the interests of all parties being considered? Is
there a greater national stake in standards that requires government involvement? This
thesis has concerned itself with a narrower variant of the problem: is there persistent
underinvestment in standards development within industries, and if so, how can it be dealt
,with?
From the viewpoint of the government and the nation, there is an increasing
national stake in standards. The timely development of good standards promotes
efficiency. The adoption of national standards by the international community can be a
source of real competitive advantage. Nations are increasingly using standards
strategically. Standards play an important role in promoting safety, public health, and
product quality. The nature of standards processes and high transaction costs make it hard
for individuals and consumers to participate directly in developing standards. For all of
these reasons, there is a felt need for a national role in improving standards processes for
the good of the country.
Individual stakeholders in specific standards have an even more direct interest in
effective standards processes. These stakeholders include manufacturers, distributors,
176For
further discussion of the high-resolution system issue, see Chapter 4.5, supra. Machine tool
examples include the conversion to metric and loss of the numerically controlled (NC) tool market: see
Chapter 4.4, supra. Data communications includes the controversy over OSI versus TCP/IP network
protocols. See Chapter 4.5 supra.
sellers, buyers, and the citizens whose lives are ultimately touched by affected products
and processes. For these players the presence or absence of effective standards
mechanisms has a direct impact, and they have a vested interest in maintaining the
standards infrastructure -- even though they may not be contributing at an optimal level to
the support of that infrastructure.'
77
5.2. The Policy Debate: Private or Public?
Current policy, within both the public and private sector, emphasizes reliance on
market mechanisms. Within the government, this policy is most clearly articulated inthe
Executive Branch directive OMB Circular A-1 19, which requires federal agencies to rely
almost exclusively on standards developed within the private sector."17 While the
government has an inordinately large impact on standards in its roles as a large purchaser
of goods and services and as a regulator, this impact is not highly coordinated or focused
beyond the specific goals of individual agencies.
79
The effect of this non-policy policy is
to support the current system, which relies on ANSI and the other major standards
organizations.
Industry policy mirrors the government's. Most standards development today
continues to take place under the imprimatur of large consensus standards committees, or
1 77
Collective action analysis implies that many actors will contribute far less than the standards
mechanisms are worth to them, largely because: benefits of their contributions accrue to many others;
the presence of the collective good may not depend on their contribution; and other players may already
have contributed to provide all of the collective good the individual is willing to pay for.
'78U.S. Office of Management and Budget, CircularNo. A-119, "Federal Participation in the
Development and Use of Voluntary Standards" (Washington, D.C.: GPO, 1982) 3.
179 For
example, NISTs information system representatives consider themselves representatives of the
governments' computer users rather than national leaders in standards setting, or even coordinators of
all government positions on standards. Burroughs, NIST, interview with OTA staff, 19 July 1991.
Similarly, the Department of Defense exercises great market power in areas such as advanced material
specifications or aerospace standards but does not act with any special responsibility for advancing a
"national interest" in standards.
de facto consortia established by industry.' 80 There are few, if any, major "grass-roots"
efforts under way to change the standards setting process within the private sector.
The debate over these policy reforms mirrors the debate in American society over
the relative merits of market mechanisms and public institutions. The market is thought to
promote economic efficiency as well as political and economic stability and the freedom to
dictate that concentration of power in the hands of the government should be
minimized. 181 It is further argued that the reliance on private institutions is engrained in the
American character, and to look to the government to make certain basic decisions about
standards for doing business, arguably issues that are inherently individual preferences,
goes against the national psyche.' 8 2
The major proponents of the current system include ANSI, ASTM, many
successful users of the consensus system, as well as individual in business and government
who philosophically favor market mechanisms over public/government action. They have
argued quite vigorously in Congressional hearings, in the trade press, and in public
appearances, that standards should be left to the private sector and any actions taken
should further emphasize the market approach.' 83
In the context of standards, the arguments in favor of private sector leadership
tend to fall into one of two categories: those that maintain the market mechanisms are
180These
consortia appear to be on the rise; for example, the many computer industry collaborations to
create standards, which are often a weigh station to official (ANSI-approved) standards. See OTA,
GlobalStandards, 51.
181For
a very cogent summary of these arguments, see Milton Friedman, Capitalismand Freedom
(Chicago: University of Chicago Press, 1982), 1-21.
182Tocqueville,
183For
as cited in OTA, Global Standards, fn 41, 46.
example, ANSI argues that the government should take greater pains to support ANSI's role as the
national standards organization and the U.S. representative to international standards organizations.
"U.S. Participation in International Standards Activities," Statement of the American National
Standards Institute before the Department of Commerce, April 3, 1990.
working very well, and those who believe that "the market is the worst system available,
except for all of the others."184 Specific arguments include the following:' 85
I.
The private sector is best suited for complex standards making -- The best way to
accurately gauge the changing needs of industry is through the "invisible hand" of
the market. Further, the creation of good standards requires the input of experts
who will actually use the standard. This expertise is more typically found in the
private sector rather than the government. Thus, it is argued, American standards
are technologically superior and better suited to actual use than their European
counterparts developed by governmental organizations. 186
2.
The private sector is currently doing a good job -- Members of the standards
community in particular point to: international successes, such as those of ASTM
or the medical devices community; the developed standards structures in industries
such as telecommunications; and the technical difficulties facing the European
community standards organizations, who are seeking to reform themselves more in
line with their American counterparts; as evidence of the merit of the current U.S.
system.187
3.
Mistrust of government -- A counterpart to the support for the private sector is a
mistrust in the ability of government. Proponents of market solution often argue
that the government does not have the flexibility or expertise to manage standards
efforts. Moreover, many argue that the non-market nature of government tends to
promote bureaucracy and inefficiency. To some extent, these arguments also
'84"Democracy is the worst form of government, except for all of the others." Winston Churchill.
18sSee generally, Peralta, ANSI, interview with author, Washington, D.C., 25 June 1992.
816This argument may be born out by the telecommunications
case study, where the myriad activities of
Committee TI raise serious doubts as to whether any government agency could hope to match them,
both in organization and in expertise. See Section 4.2, supra.
example, the European Telecommunications Standards Institute (ETSI) has studied the workings
1g7For
of the Committee TI as a model for injecting more private sector influence in its own procedures. Ivor
Knight, interview with author.
100
reflect a more philosophical concern with government control of virtually any
element of society.'8 8
4.
Cultural consistency -- While the U.S. system is different than that found in many
other countries, it is arguably much more consistent with American culture's predilection for individual power, market mechanisms, and small government.
89
On the other side of this debate, there is a consensus among economists and other
observers of social organization that certain functions require the influence of
governments. To provide some public goods, such as national defense or public works, the
need for government action is widely acknowledged.190 In part because the externalities
and costs of coordination are so high and in part due to the changing sensibilities of the
public, government is now the provider of last resort for a growing variety of functions
ranging from health and environmental regulation to basic scientific research.191 Finally,
there is a developing consensus that a more activist industrial policy is necessary to
compete in an emerging global marketplace.
192
The supporters of government involvement include standards users for whom the
system has not worked well, disaffected members of the major standards organizations,
and reform-minded observers of the system who often contrast it with interventionist
European systems. These proponents tend to argue for a greater government role in
standards-setting, including funding for standards, a government sponsored national
188Mansfield,
508. Friedman, 7-11.
89
1 See OTA, Global Standards, 20.
19°Friedman, 34-36; Mansfield, 493-495.
191See, for example, the range of activities proposed inPresidential candidate Bill Clinton's populist
campaign blueprint for changing America. Bill Clinton and Albert Gore, PuttingPeople First(New
York: Times Books, 1992).
192 Robert
Reich, The Work of Nations, (New York: Vintage Books, 1992). See generally.
101
9 1
standards organization such as those found in Europe or Canada, and/or greater
government coordination of standards efforts including mandatory use of standards.19 3
The arguments for public action reflect these concerns over market failures as well
as the appropriate role of government in safeguarding the public interest. Some of these
arguments include:
Market mechanisms are not doing the job - There is a growing sense that the
private sector is not providing sufficient levels of standards in a timely manner. The
experiences in such areas as machine tools and data communications bear
testimony to this.
Timeliness - The sluggish response of private sector standards mechanisms, as
contrasted with government led standards, is one of the primary complaints
regarding the current system. The battles over data communication protocols or
VCR formats are examples of the divisiveness and expense of long, slow standards
developments. The cumbersome committees and consensus processes of the
voluntary standards process often take years to develop new standards.
94
There is
a growing sentiment that the benefits of grass-roots standards may not outweigh
the competitive losses of late entry in the global standards arena.195
Not all stakeholders are represented - Very often all of the interested parties in a
given standard are not represented in the standards process. This is particularly
true for small actors, such as individual consumers. The cost of participation in the
process is often greater than the actual benefit to the individual stakeholder will be.
193 Such as
the Standards Council of the United States of America (SCUSA) plan. "U.S. Participation",
April 3, 1990.
194For
example, time to standard is an especially acute problem in computer standards (ANSI committee
X3), where short product life-cycles are short and the standards mechanisms remain agonizingly slow.
Lyman Chapan, interview with author, Cambridge, MA, 9 May 1991.
195For example,
the U.S. narrowly escaped Japanese preemption in the field of HDTV standards in 1988.
102
National stake in standards - There is growing recognition of the fact that
development of standards is important if the United States is to remain competitive
in the international market; the use of standards by other countries to erect trade
barriers and create their own trade arenas is illustrative of the power of standards.
Regulatory reliance on standards - Standards are increasingly used as a tool for
regulation in a variety of areas from public health and safety to the environment.
The government's interest in proper regulation confers some obligation to take a
more active role, particularly in light of the limitations noted above.
In reality, the polar paradigms offered for reforming the standards process in the
United States are more accurately portrayed as a spectrum of policy alternatives, ranging
from greater private sector autonomy at one end to complete governmental control at the
other. Some examples of the policy options available along this spectrum can be found in
OTA's 1992 Global Standards report, and include the following:196
1. Encourage greater awareness and appreciation of standards within the private
sector, through public education and other agency activities.
2. Fund standards research and education
3. Provide suppolrt for a national standards information infrastructure, such as an
electronic standards database and network to facilitate distribution as well as
participation in the development process.
4. Strengthen coordination of government standards activity by supporting and
clarifying the mission of the weakened Interagency Standards Policy Committee.
5. Provide funding for U.S. standards activities, ranging from tax credits for
standards investments to outright financing of certain standards activities.
1960OTA, Global Standards, 21-35.
103
6. Establish a public corporation for standards, by funding a new quasi-government
instrumentality, along the lines of the Tennessee Valley Authority, COMSET, or
the Corporation for Public Broadcasting, to serve as the primary U.S. standards
body.
7. Make a Federal agency responsible for coordinating and monitoring standards
activities in the U.S.
The collective action analysis in this thesis is intended to shed light on the relative merits
and drawbacks of these two approaches to standards development.
5.3. Implications of Collective Action Theory
The implications of the collective goods analysis here constitute a group of general
findings regarding --
1.
the failure of market mechanisms to provide adequate investment in standards
mechanisms;
2.
the importance of "standards infrastructures," at the industry level; and
3.
remedies available through government intervention and other external
incentives or threats.
Collective action analysis predicts that in some cases market mechanisms will not
be sufficient to provide optimal levels of standards. The model proposed and supported
here implies that in some cases, for privileged groups or some intermediate groups, market
mechanisms alone will not always provide a needed standards infrastructure. For example,
in the agriculture case study private sector initiatives proved completely incapable of
meeting the national need for standards. Similarly, in the machine tools industry reliance
on the market has led to serious underinvestment in standards processes, resulting in real
104
losses in competitive advantage and an unfair burden on the few large tool makers who
did invest in standards.
To the extent this hypothesis is true, it has obvious implications for the privatesector approach to standards setting. According to collective action theory, that approach
will not provide efficient quantities of standards unless there are other incentives, or
coercive forces, at work.
Building effective industry-wide standards "infrastructures" should be a focus of
standards policy and research. Today, most policy research focuses on individual
standards or national processes. This is where most of the motivating anecdotal evidence
for reform comes from. But at the heart of the standards system are the standards
infrastructures, the established organizations: that develop and disseminate standards; that
are looked to by industry to provide standards; that have established committees and
working groups; and which have assembled the requisite expertise from industry.
Current policy and research does not pay enough attention to standards
mechanisms at the industry level. Different industries appear to have different levels of
success at developing standards, maintaining standards-setting mechanisms, and using
standards to their competitive advantage in the international arena. The bulk of standards
setting mechanisms exist at the industry level, with industry trade and professional
associations playing a leading role and organizing and coordinating development activities;
however, most of the policy options under discussion tend to emphasize national systemic
changes.
Government involvement may be an appropriate remedy to the collective goods
dilemma. For some groups, external forces will be the only method of providing efficient
levels of standards. One such force is government intervention. Where the broad collective
interest calls for standards, the government may be able to act even when individuals will
not. This is of particular importance in areas where there is a public regulatory interest or
105
a lack of public representation, such as in consumer product safety standards or OSHA
workplace safety standards.
One of the main arguments against such intervention is that the government does
not have sufficient expertise to run a major standards program. However, as evidenced by
the USDA's position in setting agriculture standards, in some cases the government
appears able to organize itself effectively and provide standards. In the case of the USDA,
the agency has managed to both support its own efforts and maintain worldwide
leadership in agricultural standards technology.
Government involvement need not take the form of direct intervention in the
standards process. For example, in the medical devices case one reason the industry' paid
so much attention to standards was merely the threat of government regulation. Similarly,
in areas such as data communications (the GOSIP protocol) or advanced materials (for the
Defense Department) the government in its role as prominent buyer has forced industry to
invest in standards-setting without direct intervention in the market.
Other non-governmental incentives and coercive influences can be used to improve
standards processes. Government is only one means of providing incentives (through
direct support) or coercion (through penalties). The standards mechanisms themselves can
be constructed to confer extra benefits to participants, or penalties for those who do not
contribute. For example, in the technologically intensive field of telecommunications
standards, the ability to glean new information from standards meetings provides a
powerful incentive for participation, completely independent of whether or not a standard
is developed. Participants may also gain advantage through early access to standards
before they are officially published. Similarly, non-participants may suffer penalties in
terms of limited access to or higher prices for published standards.
The incentives used need not be directly related to standards development.
Support for the American Dental Association and the National Machine Tool Builders
Association standards come from member dues; however, members see the primary
106
benefits of their dues as access to trade shows, industry publications, and group lobbying
efforts, not support for standards initiatives.
The incentives or penalties used to promote participation can actually act to reduce
the collective nature of the standards infrastructure. For example, there is a benefit
conferred to participants in their ability to influence standards to their advantage. This
benefit of the standards process becomes a benefit to the individual, not to the group as a
whole, and thus reduces the collective nature of the good. Such appropriation of benefits
may serve as a powerful force to support standards development, but there is a danger as
well; privatizing the collective good may remove many of the efficiencies owed to its very
collectiveness. •97
Some other factors not modeled may play an important role in the success of
standards mechanisms. The collective goods model classifies industries based on their
size, structure, and concentration, and the importance of standards within the industry.
Other factors have been identified in the cases that may play some role in the provision of
standards. These include:
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
Competitive position and profitability of the U.S. industry, at home and abroad
Individual personalities and leadership within the industry
The presence of multinational corporations
Intersecting regulatory interests of the government
The threat of regulation
The impact of government participation as a group member
International treaties
Inspection and certification issues
Education of users - understanding the importance of standards
Professional ethics and public trust within the industry
5.4. Areas for Further Research
197 Samuelson, PublicExpenditures. Also see the discussion in Chapter 3, supra.
107
While the cases presented support the proposed model for standards development,
it is less clear how broadly useful that model will be. There are many gaps in the model;
the effects of unmodeled factors are unclear; and the applicability of the model outside of
the sample set is not proven. Among the many questions and pathways for future inquiry
are these:
1.
Is the collective action model applicable to other industries? - The collective action
hypothesis is supported by observations in only four industries. At that, the case
study observations are inherently subjective. They are based largely on discussions
and interviews with individuals within the observed system. As such, there is
clearly room for bias or sampling error. A broader study, encompassing a larger
number cf industries and relying on more quantitative measures, could be an
interesting test of the applicability of the work done here.
2.
How important is the collective action model in describing industry behavior? Are
other factors also important? - As the cases described indicate, there are a number
of other factors at work in standards development besides collective action.
Government regulation, individual personalities and leadership, industry
profitability, and user education all play a role in the development of industry
standards mechanisms. Further research could yield a more complete list of these
factors and a more complex model of the role they play. In particular, further
investigation is needed to determine how important these factors are relative to the
collective goods model.
3.
What are the implications of other theoretical writings about collective action
within groups? - A whole body of social choice and collective action literature
exists today that considers the dynamics of interest groups and group theory from
a more (sociological?) perspective. This school of thought may have some
important insights into understanding group behavior in standards setting.
108
4.
What is 'success' in standards setting? - One of the great difficulties in any
discussion of standards is defining what success is, and what the goals of the
standards process are. Existing literature often focuses on arriving at the "best"
standard, but there are many problems with this approach. A standard may be
"right" from a technical perspective, but wrong from an economic, regulatory, or
nationalistic perspective. Market reactions to standards are difficult to predict.
There is little indication of what technical advances may be inspired, or stymied, by
adoption of a standard that is good or bad, prematu re or overdue. This study has
finessed the issue to some extent by looking at a simplified, and subjective, metric:
the presence or absence of a developed industry "standards infrastructure." While
this is a useful indicator of the health of an industry's standards (see Chapter 2), a
more well understood, quantifiable measure would be a great contribution to the
study of standards.
5.
How important are standards? - One studying the role of standards in the modem
economy cannot help but be struck by the dissonance between the importance of
standards and the lack of attention and publicity they receive. Standards are not
glamour items, but half of the articles in the Wall Street Journalbusiness section
are actually about standards. What is most seriously lacking is any methodology
for understanding and quantifying the impact and importance of standards for
industry and the nation as a whole. This will be especially important as we enter an
era of "flexible specialization" where process standards coordinate activities in the
tightly coupled world of lean production.198
5.5. Conclusion
198Piore
and Sable, The Second IndustrialDivide (Basic Books, Inc.: New York, 1984). Linda Garcia, of
the Office of Technology Assessment, has spoken on the growing role for process standards in the new
manufacturing methodologies.
109
This thesis is about two related things; it is about how we can manage and
coordinate our increasingly unmanageable and unfathomable world through standards, and
it is also about how people get together to do things in their common interests.
We live in a changing, complex world. Standards provide one way to manage,
maintain, and organize that world. They serve as a sort of short-hand, a way of insuring
that a product or process contains certain features. The problem is that industry is
underinvesting in the standards infrastructures used to develop this useful shorthand. The
collective benefit from standards may actually doom them to failure; as the case studies
indicate, in some cases the only way to get needed standards mechanisms is through forces
outside of the market, whether government or otherwise.
The broader lesson to be learned here is that the invisible hand of the market is
often insufficient to supply things for the common good. Such goods often require
organization, appreciation, and enlightenment to ensure they will appear in the quantities
needed. In the context of standards, this calls for extra-market forces. These forces need
not be governmental; an organized private sector can be structured to provide other
incentives for participation. But the broader analogy should not be lost; in difficult days
and complex times, participation in the coordinated economy will increasingly require
collective action, and collective leadership, to meet the demands of the new world.
110
6.
References
Standards
Adams, Walter and James W. Brock. "Integrated Monopoly and Market Power: System
Selling, Compatibility Standards, and Market Control." QuarterlyReview of
Economics and Business 22(4) (Winter 1982): 29-42.
Berg, Sanford V. "Technical Standards as Public Goods: Demand Incentives for
Cooperative Behavior." Public Finance Quarterly 17 (January 1989): 29-54.
Besen, Stanley and Leland Johnson. CompatibilityStandards, Competition, and
Innovation in the BroadcastingIndustry. Washington, D.C.: RAND Corporation,
1986.
Braunstein, Yale M. and Lawrence J.White. "Setting Technical Compatibility Standards:
An Economic Analysis." The Antitrust Bulletin 30 (Summer 1985): 337-355.
Crane, Rhonda J. The PoliticsofInternationalStandardization:Franceand the Color TV
War. Norwood, N.J.: Ablex Publishing, 1979.
David, Paul A. "Some New Standards for the Economics of Standardization in the
Information Age." In Economic Policy and TechnologicalPerformance, edited by
Partha Dasgupta and Paul Stoneman. Cambridge: Cambridge University Press, 1987.
206-239.
Farrell, Joseph and Garth Saloner. "Installed Base and Compatibility: Innovation, Product
Preannouncements, and Predation." The American Economic Review 76 (December
1986): 940-955.
Farrell, Joseph and Garth Saloner. "Standardization, Compatibility, and Innovation."
RAND Journalof Economics 16 (Spring 1985): 70-83.
Farrell, Joseph and Carl Shapiro. "Dynamic Competition With Switching Costs." RAND
Journalof Economics 19 (Spring 1988): 123-137.
Hawkins, Richard W. "Standards and the Art of Fortress Building: Regional Power
Structures and Technical Regulations for Telecommunication." Proceedingsof the
41st InternationalTelecommunicationsAssociation Conference, 23-27 May 1991.
Chicago, IL: n.p.
Hemenway, David. Industrywide Voluntary ProductStandards. Cambridge, Mass.:
Ballinger Publishing Co., 1975.
Katz, Michael L. and Carl Shapiro. "Network Externalities, Competition, and
Compatibility." American Economic Review 75(3) (1985): 424-440.
Knapp, Julius. Office of Engineering and Technology. Federal Communications
Commission. Interview with author. Washington, DC, 8 July 1991.
Kindleberger, Charles P. "Standards as Public, Collective and Private Goods." Kyklos 36
(1983): 377-396.
111
Lecraw, Donald J. "Some Effects of Standards." Applied Economicsl6 (1984): 507-522.
Link, Albert. "Market Structure and Voluntary Product Standards." Applied Economics
15 (1983): 393-401.
Peralta, Manuel. American National Standards Institute. Interview with author. New
York, NY, 25 June 1992.
Sirbu, Marvin and Steven Stewart. Market Structure and the Emergence of Standards:
Test in the Modem Market WP-8. MIT Research Program on Communications Policy,
June 1986.
Sirbu, Marvin and Laurence E. Zwimpfer. "Standards Setting for Computer
Communication: The Case of X.25." IEEE CommunicationsMagazine 23(3) (1985):
35-45.
Statement of the American National Standards Institute before the Department of
Commerce. "U.S. Participation in International Standards Activities," April 3, 1990.
United States Congress. Office of Management and Budget. OMB CircularA-119.
"Federal Participation in the Development of Voluntary Standards." Washington,
D.C.: GPO, 1982.
United States Congress. Office of Technology Assessment. Global Standards: Building
Blocks for the Future. Washington, D.C.: GPO, March, 1992.
United States Federal Trade Commission. Bureau of Consumer Protection. Standardsand
Certification:ProposedRule and Staff Report. Washington, D.C.: GPO, 1978.
Weiss, Martin and Marvin Sirbu. "Technological Choice in Voluntary Standards
Committees: An Empirical Analysis." Economic Innovation andNew Technology,
Volume I, 111-133. Great Britain: Harwood Academic Publishers.
Collective Action, Public Choice and Industrial Organization
Arrow, Kenneth J.Social Choice and the Individual. New York. Wiley, 1951.
Blaug, Mark. Economic Theory in Retrospect. Cambridge: Cambridge University Press,
1985.
Clinton, William and Albert Gore. PuttingPeople First: How We Can All Change
America. New York: Times Books, 1992.
Friedman, Milton. Capitalismand Freedom. Chicago: University of Chicago Press, 1982.
Mansfield, Edwin. Microeconomics: Theory andApplications, 5th ed. New York: W.W.
Norton & Company, 1985.
Olson, Mancur. The Logic of Collective Action: Public Goods and the Theory of Groups.
Cambridge, Mass.: Harvard University Press, 1971.
Piore, Michael J. and Charles F. Sabel. The Second IndustrialDivide. New York: Basic
Books, Inc., Publishers, 1984.
112
Rawls, John. A Theory ofJustice. Cambridge, Mass.: Harvard University Press, 1971.
Reich, Robert. The Work ofNations. New York: Vintage Books, 1992.
Samuelson, Paul A. "The Theory of Public Expenditure." The Review of Economics and
Statistics 36 (1954): 387-389.
Sandmo, Agnar. "Public Goods." In The New Palgrave:A DictionaryofEconomics,
edited by J. Eatwell, M. Milgate, and P. Newman. London: Macmillan, 1987. 10611065.
Sen, Amartya. "Social Choice." In The New Palgrave:A DictionaryofEconomics, edited
by J. Eatwell, M. Milgate, and P. Newman. London: Macmillan, 1987. 1040-1044.
Tullock, Gordon. "Public Choice." In The New Palgrave:A DictionaryofEconomics,
edited by J. Eatwell, M. Milgate, and P. Newman. London: Macmillan, 1987. 10401044.
Weber, Max. The ProtestantEthic andthe Spirit of Capitalism. Trans. Talcott Parsons.
New York: Charles Scribner's Sons, 1958.
Agriculture
Beniger, James R. The ControlRevolution: Technology and the Economic Originsof
the Information Society. Cambridge, MA: Harvard University Press, 1986.
Bomer, Sharon. Assistant to the Director, Fruit and Vegetable Division, U.S. Department
of Agriculture. Interview with author. Washington, D.C., 22 July 1991.
Clayton, Kenneth. Deputy Administrator, Agricultural Marketing Service, U.S.
Department of Agriculture. Letter to author. 10 July 1991.
Danhof, Clarence H. Change in Agriculture: The Northern UnitedStates, 1820-1870.
Cambridge, MA: Harvard University Press, 1969.
Mullins, Fred. Cotton Division, U.S. Department of Agriculture. Interview with author.
Washington, D.C., 22 July 1991.
U.S. Department of Agriculture. Agricultural Marketing Service. AgriculturalMarketing
Service InternationalPrograms.Washington, D.C.: GPO, 1991.
U.S. Department of Agriculture. Agricultural Marketing Service. "Core Speech: Fresh
Products Branch." Photocopy of typescript.
U.S. Department of Agriculture. Agricultural Marketing Service. This is AMS.
Washington, D.C.: GPO, 1991.
Medical Devices
Cornish, Steve. American National Standards Institute. Telephone interview with OTA
staff. 28 March 1991.
113
Cornish, Steve. American National Standards Institute. Letter to Office of Technology
Assessment. 9 July 1991.
Kimmelman, Gene. The Public Citizen. Interview with author. Washington, D.C., 22 July
1991.
Liebler, Bernard. Director, Health Industry Manufacturers Association. Telephone
interview with author. 10 July 1992.
Miller, Mike. Executive Director, Association for the Advancement of Medical
Instrumentation. Telephone interview with author. 14 July 1991.
Miller, Mike. Executive Director, Association for the Advancement of Medical
Instrumentation. Telephone interview with author. 13 August 1991.
Miller, Mike. Executive Director, Association for the Advancement of Medical
Instrumentation. Letter to author. 14 August 1991.
Stanford, Sharon. American Dental Association. Telephone interview with author. 11 July
1991.
Stanford, Sharon. American Dental Association. Letter to author. 15 July 1991.
Troje, Suzanne. Office of the U.S. Trade Representative. Interview with author.
Washington, D.C., 12 June 1991.
Troje, Suzanne. Office of the U.S. Trade Representative. Telephone interview with
author. 12 June 1991.
U.S. Congress. Office of Technology Assessment. FederalPoliciesand the Medical
Devices Industry. Washington, D.C.: GPO, 1984.
Willingmeyer, George. American National Standards Institute. Interview with author.
New York, N.Y., 24 June 1991.
Wolfe, Sid. The Public Citizen. Telephone interview with author. 7 July 1991.
Telecommunications and Information Technology
Berg, J.L. and H. Schunny. An Analysis of the Information Technology Standardization
Process. North-Holland, Amsterdam, 1989.
Buzacott, Alan. BroadbandTelecommunications Standards.Master's Thesis, MIT.
Cambridge, MA, May 1990.
Cargill, Carl F. Information Technology Standardization: Theory, Process, and
Organizations. Digital Press: 1989.
Chapan, Lyman. Interview with author. Cambridge, MA, 9 May 1992.
Committee TI-Telecommunications, Annual Report, 1990.
Drake, William J. and Lee McKnight. "Telecommunications Standards in the Global
Information Economy: The Impact of Deregulation and Commercialization."
Promethee, March 1988.
114
Johnson, Philip. AT&T Bell Laboratories. Interview with author. San Francisco, CA, 6
August 1991.
Katz, Michael L. and Carl Shapiro. "Technical Adoption in the Presence of Network
Externalities." Journalof PoliticalEconomy, 94(4), 1986.
Lerner, E., "Microcomputer Standards: Weighing the Pros and Cons." IEEE Spectrum,
May 1981.
Lifchus, Ian. "Standards Committee Ti-Telecommunications." IEEE Communications
Magazine, January, 1985.
Lohr, Steve. "Computer Makers Told to Get Involved in Rules." New York Times, 23
February 1993.
McKnight, Lee. "International Telecommunications and Competitiveness." Promethee.
McNamara, W.J., III. BellSouth Services. Interview with author. San Francisco, CA, 5
August 1991.
Pool, Ithiel de Sola Pool. Technologies of Freedom. Cambridge, Mass.: The Belknap
Press, 1983.
Reilly, Arthur. Vice Chairman, Committee TI-Telecommunications. Interview with
author. San Francisco, CA, 5 August 1991.
Smoot, Oliver. Vice President, CBEMA. Interview with author. Washington, D.C., 12
August 1991.
Solomon, Richard. Telephone interview with author. 27 June 1991.
Solomon, Richard. "What Happened After Bell Spilled the Acid?" Telecommunications
Policy, June, 1978.
Stampleman, Joseph. Apple Computer. Telephone interview with author. 28 February
1993.
United States Congress. Office of Technology Assessment. CriticalConnections:
Communicationsfor the Future. Washington, D.C.: GPO, 1990.
United States Department of State. Interview with OTA Staff. Washington, D.C., 12
June 1991.
Varnassi, Michael. MCI. Interview with author. San Francisco, CA, 5 August 1991.
Machine Tools
Alic, John. Office of Technology Assessment. Interview with author. Washington, D.C.,
8 July 1991.
Bratkovich, Anthony. National Machine Tool Builders Association. Interview with
author. McLean, VA, 10 July 1992.
Carlsson, Charles. National Machine Tool Builders Association. Interview with author.
McLean, VA, 10 July 1992
115
Dertouzos, Michael L., Richard K. Lester and Robert M. Solow. Made In America:
Regainingthe Reproductive Fdge. MIT Press, 1989.
"Machine Tools: America's Muscles." McLean, VA: National Machine Tool Builders
Association, 1981.
Made In America: Working Papersof the MIT Commission on IndustrialProductivity,
MIT Press, 1989.
116